Wersja 6 Visual Basica (zwana również dla kontrastu z VB.NET) VB.COM ma szansę na przeżycie. Analitycy zwracają uwagę na analogię z firmą Coca-Cola, która również kiedyś miała kłopoty z nowym produktem (jego akceptacją) i była zmuszona zachować stary smak w postaci Classic-Cola. W przypadku VB.NET nie jest to ulepszona wersja VB 6 ale całkowicie nowa jakość wprowadzajaca nie tylko pełne OO ale również podporządkowana zasadom uruchomienia aplikacji w ramach .NET Framework i CLR. Dlatego MS ma prawo na ten ruch.
Wskazuje się na zasługi jakie ten prosty język VB 6 zrobił dla upowszechnienia platformy grubego klienta (rich client platform). Mimo, że po zaprzestaniu rozwoju VB 6 w 2001 i wprowadzeniu Visual Basic .NET firma MS zaoferowała ścieżki migracyjne, jednak przeniesienie aplikacji z VB 6 na nową platformę jest niezwykle trudne. Petycja proponuje zapewnienie rozwoju VB 6 oraz VBA (Basic for Application) dalszej, rownoległej do paltformy .NET ścieżki rozwoju uwzględniającej masowe zapotrzebowanie użytkowników. Nowa wersja VB 6 nosiłaby nazwę VB.COM w celu podkreślenia, że bazuje na technologii COM i współdziałałaby z .NET Visual Studio.
czwartek, marca 10, 2005
CDN losów VB.COM
Wersja 6 Visual Basica (zwana również dla kontrastu z VB.NET) VB.COM ma szansę na przeżycie. Analitycy zwracają uwagę na analogię z firmą Coca-Cola, która również kiedyś miała kłopoty z nowym produktem (jego akceptacją) i była zmuszona zachować stary smak w postaci Classic-Cola. W przypadku VB.NET nie jest to ulepszona wersja VB 6 ale całkowicie nowa jakość wprowadzajaca nie tylko pełne OO ale również podporządkowana zasadom uruchomienia aplikacji w ramach .NET Framework i CLR. Dlatego MS ma prawo na ten ruch.
Wskazuje się na zasługi jakie ten prosty język VB 6 zrobił dla upowszechnienia platformy grubego klienta (rich client platform). Mimo, że po zaprzestaniu rozwoju VB 6 w 2001 i wprowadzeniu Visual Basic .NET firma MS zaoferowała ścieżki migracyjne, jednak przeniesienie aplikacji z VB 6 na nową platformę jest niezwykle trudne. Petycja proponuje zapewnienie rozwoju VB 6 oraz VBA (Basic for Application) dalszej, rownoległej do paltformy .NET ścieżki rozwoju uwzględniającej masowe zapotrzebowanie użytkowników. Nowa wersja VB 6 nosiłaby nazwę VB.COM w celu podkreślenia, że bazuje na technologii COM i współdziałałaby z .NET Visual Studio.
Wskazuje się na zasługi jakie ten prosty język VB 6 zrobił dla upowszechnienia platformy grubego klienta (rich client platform). Mimo, że po zaprzestaniu rozwoju VB 6 w 2001 i wprowadzeniu Visual Basic .NET firma MS zaoferowała ścieżki migracyjne, jednak przeniesienie aplikacji z VB 6 na nową platformę jest niezwykle trudne. Petycja proponuje zapewnienie rozwoju VB 6 oraz VBA (Basic for Application) dalszej, rownoległej do paltformy .NET ścieżki rozwoju uwzględniającej masowe zapotrzebowanie użytkowników. Nowa wersja VB 6 nosiłaby nazwę VB.COM w celu podkreślenia, że bazuje na technologii COM i współdziałałaby z .NET Visual Studio.
środa, marca 09, 2005
Bardzo ciekawe linki z www.windows2003.pl
- Rezygnacja z .NET: http://www.ddj.com/documents/s=9211/ddj050201dnn/
- Co warto czytać: http://www.microsoft.com/poland/developer/press/default.htm
- Office 2003 i XML i Smart Client: http://www.windows2003.pl/news.aspx?pn=2&order=&cat=D#
- Ciekawy blog: http://www.michal.osmenda.com/
- Przechwytywanie Webcast-ów: http://msdn.microsoft.com/SQL/WebcastDownloadInstructions.aspx
- Szukamy blogów: http://www.windows2003.pl/news.aspx?cat=M&id=1811
- Inny ciekawy blog: http://www.windows2003.pl/communities.aspx?cat=4
- Jeszcze jeden: http://www.howto.blog.pl/
- WS: http://www.news2news.com/vfp/?article=5
- WS po raz drugi: http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnfoxtk03/html/ft03g.asp
- Sysinternals: http://www.sysinternals.com/ntw2k/freeware/psservice.shtml
Ciekawe adresy JavaScript
The following additional Internet resources may help you as you enhance your Web site.
Wiadomości dnia
- Netscape 8 Beta: Powerful But Buggy.Netscape is the Elvis of browsers. It started out lean, sexy, and breathtaking. But as it aged, it became a bloated, lethargic joke. With Netscape 8, released in beta on Thursday, the King is back.If you can't wait, the browser is a free download from Netscape, which is a business unit of America Online, which in turn is part of Time Warner. (na podstawie)
- Pod koniec marca kończy się support dla Visual Basic 6.0, oznacza to brak krytycznych poprawek i aktualnień SP. Język ten rozwijany jest od 1999r. Do firmy MS wysłano petycję z prośbą o przedłużenie okresu bezpłatnej opieki (opieka płatna będzie świadvzona do 2008). Petycję podpisało ok 130 developerów (MVP m.in. Rich Levin) wskazując na podobny precedens (C++ jest nadal wspierany mimo promowania C#).
wtorek, marca 08, 2005
Jak zatrudniać
Everyone thinks they're hiring the top 1%.
Martin Fowler: “We are still working hard to hire only the very top fraction of software developers (the target is around the top 0.5 to 1%).”
Me: “We get between 100 and 200 [resumes] per opening.”
I remember when I started working for David Shaw he told us they only hired "1 out of 200."
I hear this from almost every software company. "We hire the top 1% or less," they all say.
Could they all be hiring the top 1%? Where are all the other 99%? General Motors?
I had an insight the other day.
Quiz: If you get 200 resumes, and you hire 1 person, are you hiring the top 0.5% of software developers?
"No," you say, "your screening process is unlikely to find the best person out of 200."
Agreed. OK. Let's say you had a magical screening process that actually allowed you to find the "best" person.
"No," you say, "people are good at different things. There's no absolute, forced ranking of developers that makes sense."
Agreed. Let's simplify for the moment and assume that all software developers in the world could be ranked in absolute order of skill, and that you had a magical screening process that found the "best" person from any field.
Now, when you get those 200 resumes, and hire the best person from the top 200, does that mean you're hiring the top 0.5%?
"Maybe."
No. You're not. Think about what happens to the other 199 that you didn't hire.
They go look for another job.
That means, in this horribly simplified universe, that the entire world could consist of 1,000,000 programmers, of whom the worst 199 keep applying for every job and never getting them, but the best 999,801 always get jobs as soon as they apply for one. So every time a job is listed the 199 losers apply, as usual, and one guy from the pool of 999,801 applies, and he gets the job, of course, because he's the best, and now, in this contrived example, every employer thinks they're getting the top 0.5% when they're actually getting the top 99.9801%.
The top 0.5% usually have jobs. They have jobs where they do very well, so their employers pay them lots of money and do whatever it takes to keep them happy. (I know. Oversimplification. Lots of employers try to drive out the good software developers because they complain a lot and demand high salaries. Still.)
Those 200 resumes you got from Craigslist? Those consist of the one guy who happened to be good, but he's only applying for a job because his wife wants to be nearer to her family, and the usual floating population of 199 people who apply for every single job and are qualified for none. And now you think you're being "super selective" but you're not, it's just a statistical fallacy.
I'm exaggerating a lot, but the point is, when you select 1 out of 200 applicants, the other 199 don't give up and go into plumbing (although I wish they would... plumbers are impossible to find). They apply again somewhere else, and contribute to some other employer's self-delusions about how selective they are.
In fact, one thing I have noticed is that the people who I consider to be good software developers barely ever apply for jobs at all. I know lots of great people who took a summer internship on a whim and then got permanent offers. They only ever applied for one or two jobs in their lives.
On the other hand there are people out there who appear to be applying to every job on Monster.com. I'm not kidding. They spam their resume to hundreds or thousands of employers. A lot of times I can see this because there are actually hundreds of "job" aliases in the "To:" line of their email. (Some evil part of me wants to "reply-to-all" the rejection note I send them, but I usually overcome the urge).
It's pretty clear to me that just because you're hiring the top 0.5% of all applicants for a job, doesn't mean you're hiring the top 0.5% of all software developers. You could be hiring from the top 10% or the top 50% or the top 99% and it would still look, to you, like you're rejecting 199 for every 1 that you hire.
By the way, it's because of this phenomenon—the fact that many of the great people are never on the job market—that we are so aggressive about hiring summer interns. This may be the last time these kids ever show up on the open market. In fact we hunt down the smart CS students and individually beg them to apply for an internship with us, because if you wait around to see who sends you a resume, you're already missing out.
Martin Fowler: “We are still working hard to hire only the very top fraction of software developers (the target is around the top 0.5 to 1%).”
Me: “We get between 100 and 200 [resumes] per opening.”
I remember when I started working for David Shaw he told us they only hired "1 out of 200."
I hear this from almost every software company. "We hire the top 1% or less," they all say.
Could they all be hiring the top 1%? Where are all the other 99%? General Motors?
I had an insight the other day.
Quiz: If you get 200 resumes, and you hire 1 person, are you hiring the top 0.5% of software developers?
"No," you say, "your screening process is unlikely to find the best person out of 200."
Agreed. OK. Let's say you had a magical screening process that actually allowed you to find the "best" person.
"No," you say, "people are good at different things. There's no absolute, forced ranking of developers that makes sense."
Agreed. Let's simplify for the moment and assume that all software developers in the world could be ranked in absolute order of skill, and that you had a magical screening process that found the "best" person from any field.
Now, when you get those 200 resumes, and hire the best person from the top 200, does that mean you're hiring the top 0.5%?
"Maybe."
No. You're not. Think about what happens to the other 199 that you didn't hire.
They go look for another job.
That means, in this horribly simplified universe, that the entire world could consist of 1,000,000 programmers, of whom the worst 199 keep applying for every job and never getting them, but the best 999,801 always get jobs as soon as they apply for one. So every time a job is listed the 199 losers apply, as usual, and one guy from the pool of 999,801 applies, and he gets the job, of course, because he's the best, and now, in this contrived example, every employer thinks they're getting the top 0.5% when they're actually getting the top 99.9801%.
The top 0.5% usually have jobs. They have jobs where they do very well, so their employers pay them lots of money and do whatever it takes to keep them happy. (I know. Oversimplification. Lots of employers try to drive out the good software developers because they complain a lot and demand high salaries. Still.)
Those 200 resumes you got from Craigslist? Those consist of the one guy who happened to be good, but he's only applying for a job because his wife wants to be nearer to her family, and the usual floating population of 199 people who apply for every single job and are qualified for none. And now you think you're being "super selective" but you're not, it's just a statistical fallacy.
I'm exaggerating a lot, but the point is, when you select 1 out of 200 applicants, the other 199 don't give up and go into plumbing (although I wish they would... plumbers are impossible to find). They apply again somewhere else, and contribute to some other employer's self-delusions about how selective they are.
In fact, one thing I have noticed is that the people who I consider to be good software developers barely ever apply for jobs at all. I know lots of great people who took a summer internship on a whim and then got permanent offers. They only ever applied for one or two jobs in their lives.
On the other hand there are people out there who appear to be applying to every job on Monster.com. I'm not kidding. They spam their resume to hundreds or thousands of employers. A lot of times I can see this because there are actually hundreds of "job" aliases in the "To:" line of their email. (Some evil part of me wants to "reply-to-all" the rejection note I send them, but I usually overcome the urge).
It's pretty clear to me that just because you're hiring the top 0.5% of all applicants for a job, doesn't mean you're hiring the top 0.5% of all software developers. You could be hiring from the top 10% or the top 50% or the top 99% and it would still look, to you, like you're rejecting 199 for every 1 that you hire.
By the way, it's because of this phenomenon—the fact that many of the great people are never on the job market—that we are so aggressive about hiring summer interns. This may be the last time these kids ever show up on the open market. In fact we hunt down the smart CS students and individually beg them to apply for an internship with us, because if you wait around to see who sends you a resume, you're already missing out.
Ciekawy skrypt
SCRIPTING ANSWERS: Who's with Whom
By Don Jones, Contributing Editor
Here's a little trick with the Active Directory Services Interface
(ADSI) WinNT provider. And pay attention: The WinNT provider
works great with Active Directory domains, as well as local
computer user accounts and groups!
Let's say you have a user name in variable sUser, and a group
name in sGroup. You want to know if sUser is a member of
sGroup or not (a handy trick for logon scripts, for example).
Start by using ADSI to get a reference to each object:
Dim oUser, oGroup, sUser, sGroup
sUser = "Don"
sGroup = "Domain Users"
Set oUser = GetObject("WinNT://MyDomain/" & _
sUser & ",user")
Set oGroup = GetObject("WinNT://MyDomain/" & _
sGroup & ",group")
Here's the cool bit, which isn't available with Active Directory's
native LDAP provider, but which works dandy with the WinNT provider:
If oGroup.IsMember(oUser.aDSPath) Then
'He's a member
Else
'Not in the club
End If
You can also shortcut this; here's the entire example:
Dim oGroup, sGroup
sGroup = "Domain Users"
Set oGroup = GetObject("WinNT://MyDomain/" & _
sGroup & ",group")
If oGroup.IsMember(("WinNT://MyDomain/" & _
sUser & ",user") Then
'He's a member
Else
'Not in the club
End If
A quick and easy way to check group membership.
By Don Jones, Contributing Editor
Here's a little trick with the Active Directory Services Interface
(ADSI) WinNT provider. And pay attention: The WinNT provider
works great with Active Directory domains, as well as local
computer user accounts and groups!
Let's say you have a user name in variable sUser, and a group
name in sGroup. You want to know if sUser is a member of
sGroup or not (a handy trick for logon scripts, for example).
Start by using ADSI to get a reference to each object:
Dim oUser, oGroup, sUser, sGroup
sUser = "Don"
sGroup = "Domain Users"
Set oUser = GetObject("WinNT://MyDomain/" & _
sUser & ",user")
Set oGroup = GetObject("WinNT://MyDomain/" & _
sGroup & ",group")
Here's the cool bit, which isn't available with Active Directory's
native LDAP provider, but which works dandy with the WinNT provider:
If oGroup.IsMember(oUser.aDSPath) Then
'He's a member
Else
'Not in the club
End If
You can also shortcut this; here's the entire example:
Dim oGroup, sGroup
sGroup = "Domain Users"
Set oGroup = GetObject("WinNT://MyDomain/" & _
sGroup & ",group")
If oGroup.IsMember(("WinNT://MyDomain/" & _
sUser & ",user") Then
'He's a member
Else
'Not in the club
End If
A quick and easy way to check group membership.
Ciekawe użycie div do pozycjonowania elementów na stronie
********** THE TIP **********
Use nested div elements to standardize browser behavior when
seeking information, such as padding size, from JavaScript
If your JavaScript needs to know the exact position and size of
elements on the page (e.g., to create and control DHTML effects),
you may have some problems with the many variations between
browsers. For instance, to dynamically calculate the size of
paddings and borders (assuming these might change in your
application), you'll need different approaches for different browsers.
One shortcut that sometimes works to simplify these sorts of
calculations is to add extra nested
Use nested div elements to standardize browser behavior when
seeking information, such as padding size, from JavaScript
If your JavaScript needs to know the exact position and size of
elements on the page (e.g., to create and control DHTML effects),
you may have some problems with the many variations between
browsers. For instance, to dynamically calculate the size of
paddings and borders (assuming these might change in your
application), you'll need different approaches for different browsers.
One shortcut that sometimes works to simplify these sorts of
calculations is to add extra nested
tags. For instance, say
your element has padding. You can nest another
your element has padding. You can nest another
inside that
and place your content there. That way, you can use offsetLeft
and offsetTop to find the position of the inner element in
relation to the outer one instead of using the browser-specific
approaches for returning the padding value.
and place your content there. That way, you can use offsetLeft
and offsetTop to find the position of the inner element in
relation to the outer one instead of using the browser-specific
approaches for returning the padding value.
Przeszukiwanie pulpitu
Google udoskonala swoją wyszukiwarkę lokalną wraz z udostepnieniem API do wykorzystania we własnych rozwązaniach
Google updates desktop search tool
Juan Carlos Perez, IDG News Service
08/03/2005 08:46:36
Google is set to launch on Monday an updated version of its desktop search tool whose enhancements include the ability to search the full text of Adobe Systems PDF files and the metadata of multimedia files, a Google executive said.
Until now, the product could only index the names of PDF and multimedia files, such as image, music and video files. But the new version improves on this by indexing the entire content of PDF files and the metadata of multimedia files, such as song and artist names in music files, said Nikhil Bhatla, a Google product manager.
With this version, the product officially exits its beta, or test, phase and is now considered a finished product that, as is customary for all software, will continue to be regularly enhanced. "We've taken the product out of beta because now we have all the file types and features that were high on the list of user requests," Bhatla said, adding that by removing the beta tag, Google is hoping more users will feel encouraged to download and install the product.
Competition is fierce among providers of tools such as this one that let users index and retrieve information found on their PCs, a task for which few products existed until recently. Desktop search is viewed as an important area of the overall search engine market, because increasingly users expect to be able to find information on their PCs in the same way they find information on the Internet.
Although it isn't clear yet how search engine vendors will make money from these tools, most of which are free, it is generally agreed that a user who becomes loyal to a desktop search product is highly likely to extend that loyalty to the tool maker's Internet search engine. In recent years, the market for online ads that search engines serve up with their query results has exploded. However, a big challenge for search engine vendors is to find ways to foster loyalty among their users. Studies have shown users feel little attachment to particular engines.
Google introduced its desktop product in October of last year, joining Lycos and several smaller, niche players, such as Copernic Technologies, X1 Technologies and Blinkx, which already had desktop search tools available. However, Google beat big search engine providers such as Microsoft, Ask Jeeves, Yahoo and America Online, all of which subsequently released their own test versions of desktop search tools. Yahoo partnered with X1 and AOL with Copernic for their respective desktop search entries.
Other improvements in the Google desktop tool, which is free, are support for the Mozilla Foundation's Firefox browser and Thunderbird e-mail application and for America Online's Netscape browser and e-mail application. Previously, the product supported only Microsoft's Internet Explorer browser and Outlook and Outlook Express e-mail applications.
Moreover, Google is also releasing on Monday a tool kit for developers to create plug-in applications that extend the functionality of the desktop search product via open application programming interfaces (APIs.) The software development kit will be available at http://desktop.google.com, along with documentation, sample code and some plug-ins that have already been built.
One such plug-in lets users index instant messaging (IM) conversations from Cerulean Studios' popular Trillian IM application; the original version of the product only indexed IM sessions from AOL's AIM service. Another plug-in that will be available later will let users search the metadata of music files acquired from Apple Computer's iTunes music store, Bhatla said.
Soon, a third-party developer will create a plug-in that, using speech-to-text technology, will allow the product to transcribe the content of audio and video files and make it searchable, thus deepening the indexing capabilities of those files beyond metadata, Bhatla said.
Although this desktop search tool is designed for use by consumers, Google continues to work hard at developing a version that is appropriate for the workplace, he said. However, this new version does have one enterprise feature: It recognizes Microsoft group policy parameters on a PC and cancels its own installation if the parameters state that the Google desktop tool can't be run on that machine.
Finally, Google's desktop search tool will be available on Monday for the first time in Chinese and Korean, which means that the product's interface, support and documentation have been translated to those languages
Google updates desktop search tool
Juan Carlos Perez, IDG News Service
08/03/2005 08:46:36
Google is set to launch on Monday an updated version of its desktop search tool whose enhancements include the ability to search the full text of Adobe Systems PDF files and the metadata of multimedia files, a Google executive said.
Until now, the product could only index the names of PDF and multimedia files, such as image, music and video files. But the new version improves on this by indexing the entire content of PDF files and the metadata of multimedia files, such as song and artist names in music files, said Nikhil Bhatla, a Google product manager.
With this version, the product officially exits its beta, or test, phase and is now considered a finished product that, as is customary for all software, will continue to be regularly enhanced. "We've taken the product out of beta because now we have all the file types and features that were high on the list of user requests," Bhatla said, adding that by removing the beta tag, Google is hoping more users will feel encouraged to download and install the product.
Competition is fierce among providers of tools such as this one that let users index and retrieve information found on their PCs, a task for which few products existed until recently. Desktop search is viewed as an important area of the overall search engine market, because increasingly users expect to be able to find information on their PCs in the same way they find information on the Internet.
Although it isn't clear yet how search engine vendors will make money from these tools, most of which are free, it is generally agreed that a user who becomes loyal to a desktop search product is highly likely to extend that loyalty to the tool maker's Internet search engine. In recent years, the market for online ads that search engines serve up with their query results has exploded. However, a big challenge for search engine vendors is to find ways to foster loyalty among their users. Studies have shown users feel little attachment to particular engines.
Google introduced its desktop product in October of last year, joining Lycos and several smaller, niche players, such as Copernic Technologies, X1 Technologies and Blinkx, which already had desktop search tools available. However, Google beat big search engine providers such as Microsoft, Ask Jeeves, Yahoo and America Online, all of which subsequently released their own test versions of desktop search tools. Yahoo partnered with X1 and AOL with Copernic for their respective desktop search entries.
Other improvements in the Google desktop tool, which is free, are support for the Mozilla Foundation's Firefox browser and Thunderbird e-mail application and for America Online's Netscape browser and e-mail application. Previously, the product supported only Microsoft's Internet Explorer browser and Outlook and Outlook Express e-mail applications.
Moreover, Google is also releasing on Monday a tool kit for developers to create plug-in applications that extend the functionality of the desktop search product via open application programming interfaces (APIs.) The software development kit will be available at http://desktop.google.com, along with documentation, sample code and some plug-ins that have already been built.
One such plug-in lets users index instant messaging (IM) conversations from Cerulean Studios' popular Trillian IM application; the original version of the product only indexed IM sessions from AOL's AIM service. Another plug-in that will be available later will let users search the metadata of music files acquired from Apple Computer's iTunes music store, Bhatla said.
Soon, a third-party developer will create a plug-in that, using speech-to-text technology, will allow the product to transcribe the content of audio and video files and make it searchable, thus deepening the indexing capabilities of those files beyond metadata, Bhatla said.
Although this desktop search tool is designed for use by consumers, Google continues to work hard at developing a version that is appropriate for the workplace, he said. However, this new version does have one enterprise feature: It recognizes Microsoft group policy parameters on a PC and cancels its own installation if the parameters state that the Google desktop tool can't be run on that machine.
Finally, Google's desktop search tool will be available on Monday for the first time in Chinese and Korean, which means that the product's interface, support and documentation have been translated to those languages
Taksonomia - Gmail, Technorati, WinFS - cogitating reticulation
Gmail, Technorati, WinFS - cogitating reticulation ITworld.com, Ebusiness in the Enterprise 3/9/05
Sean McGrath, ITworld.com
I am ever-so-slightly synesthetic[1] when it comes to words. Take the word 'reticulation', for example. I recently added this one to my vocabulary. In my mind's eye, the word appears as a square sheet of aluminum, folded into origami patterns, painted the color of hubris, smelling of harpsichord and humming like magenta.
Actually, no, it doesn't. I made that up. Making up silly definitions of words helps me to stop forgetting them. I do not want to forget what 'reticulation' means. It is too useful a word to forget. Besides, this article could not exist without it.
Pull up a chair and I'll tell you what happened. It all started when I made a comment on my blog[2] to the effect that hierarchies are things we cannot think within, but equally cannot think without. I am not quite sure what it means but it sounded good at the time.
A fellow blogger, Hamish Harvery[3] linked to my blog and recommended Arthur Koestler's book 'The Ghost in the Machine'. I have not read it yet but I understand that it talks about social structures. It talks about how we create organizational hierarchies (arborization) and then join these hierarchies together into complex organizational networks (reticulation).
What on Earth am I talking about? Sit back down on that chair and I'll tell you. I am talking about digital information and how it is organized. I am talking about how we humans manage to navigate the infinite seas of digital information without our heads exploding.
Here is how we do it. If your head works the way most people's heads work, your first port of call in organizing raw information of any form is to put it into a hierarchy. There comes a point however, where you find that hierarchies are not enough to capture the rich structure of information. You start to join bits of hierarchies to each other in complex ways. As information hierarchies mature, they have a way of ceasing to be pure hierarchies. Arborization turns to reticulation.
We humans have been doing arborization for a long, long time. From the Dewey Decimal Classification system to Taxonomies of living things. We humans take to hierarchies like fish take to water. They combine great power with great simplicity. It helps greatly that hierarchies map naturally to the physical world of words on pieces of paper. Hierarchies have a natural ally in paper books as they can be linearized into a sequence of pages. If you have ever looked at a table of contents (say 'yes') then you know what I mean.
We humans have also been doing reticulation for a long time. However, it historically has been a more challenging information management technique as it does not map to the physical world of paper very easily. Even simple forms of reticulation like cross references are a pain in the real world. Anybody who has ever tried to read a piece of legislation and found themselves with a dozen separate books on their desk for following the complex cross-referencing knows what I mean.
Then, the web happened. A better incubator for reticulation techniques is hard to imagine. Hypertext subverts hierarchy. Actually, no. I don't think it does. I think hypertext supplements hierarchy with a basic form of reticulation.
Finally, I am getting to the piece you want to read. Interesting forms of reticulation are popping up everywhere these days. It is as if hypertext is now part of our collective consciousness and we have lost any fear of it. We are happy to use it as a base for more powerful forms of information management.
What on Earth am I talking about? Sit back down on that chair one more time and I'll tell you. I'm talking about the label system in GMail[4]. Have you seen how that works? There is no great big folder system to speak of. The folders in GMail are created on-the-fly based on the labels you attach to individual e-mails. You are no longer restricted to putting an e-mail in one place and one place only. You are no longer restricted to a hierarchical filing structure. The time honored metaphors of the filing cabinet and the shelf are no longer helpful.
I'm also talking about the faceted classification in WinFS[5]. Think of the way you organize files on a file system. Now, instead of thinking about fixing the names of folders and picking one folder to put your file in, think about that file being, effectively, in as many folders as you like, all at the same time. Last but not least, I'm talking about the emergence of Folksonomies[6] and the intriguing Technorati tagging system[7].
GMail, WinFS, Technorati, all these developments are reticulations of one form or another. It is as if overnight, the world has woken up to the power of it all. (In reality, it is I that has woken up to the power of it all.).
Metadata enthusiasts the world over must be shaking their heads in disbelief. Finally, it's happening - that's great! However, it is not really happening the way the metadata enthusiasts (or theorists) would have predicted. Welcome to the World Wide Web where the only thing that is completely predictable is its total unpredictability.
[1] http://www.apa.org/monitor/mar01/synesthesia.html
[2] http://seanmcgrath.blogspot.com/archives/
[3] http://weblog.hamishharvey.com/mishmash/2005/01/on_hierarchy.html
[4] http://www.beelerspace.com/index.php?p=806
[5] http://msdn.microsoft.com/msdnmag/issues/04/01/WinFS/default.aspx
[6] http://en.wikipedia.org/wiki/Folksonomy
[7] http://www.technorati.com/tag/
Sean McGrath, ITworld.com
I am ever-so-slightly synesthetic[1] when it comes to words. Take the word 'reticulation', for example. I recently added this one to my vocabulary. In my mind's eye, the word appears as a square sheet of aluminum, folded into origami patterns, painted the color of hubris, smelling of harpsichord and humming like magenta.
Actually, no, it doesn't. I made that up. Making up silly definitions of words helps me to stop forgetting them. I do not want to forget what 'reticulation' means. It is too useful a word to forget. Besides, this article could not exist without it.
Pull up a chair and I'll tell you what happened. It all started when I made a comment on my blog[2] to the effect that hierarchies are things we cannot think within, but equally cannot think without. I am not quite sure what it means but it sounded good at the time.
A fellow blogger, Hamish Harvery[3] linked to my blog and recommended Arthur Koestler's book 'The Ghost in the Machine'. I have not read it yet but I understand that it talks about social structures. It talks about how we create organizational hierarchies (arborization) and then join these hierarchies together into complex organizational networks (reticulation).
What on Earth am I talking about? Sit back down on that chair and I'll tell you. I am talking about digital information and how it is organized. I am talking about how we humans manage to navigate the infinite seas of digital information without our heads exploding.
Here is how we do it. If your head works the way most people's heads work, your first port of call in organizing raw information of any form is to put it into a hierarchy. There comes a point however, where you find that hierarchies are not enough to capture the rich structure of information. You start to join bits of hierarchies to each other in complex ways. As information hierarchies mature, they have a way of ceasing to be pure hierarchies. Arborization turns to reticulation.
We humans have been doing arborization for a long, long time. From the Dewey Decimal Classification system to Taxonomies of living things. We humans take to hierarchies like fish take to water. They combine great power with great simplicity. It helps greatly that hierarchies map naturally to the physical world of words on pieces of paper. Hierarchies have a natural ally in paper books as they can be linearized into a sequence of pages. If you have ever looked at a table of contents (say 'yes') then you know what I mean.
We humans have also been doing reticulation for a long time. However, it historically has been a more challenging information management technique as it does not map to the physical world of paper very easily. Even simple forms of reticulation like cross references are a pain in the real world. Anybody who has ever tried to read a piece of legislation and found themselves with a dozen separate books on their desk for following the complex cross-referencing knows what I mean.
Then, the web happened. A better incubator for reticulation techniques is hard to imagine. Hypertext subverts hierarchy. Actually, no. I don't think it does. I think hypertext supplements hierarchy with a basic form of reticulation.
Finally, I am getting to the piece you want to read. Interesting forms of reticulation are popping up everywhere these days. It is as if hypertext is now part of our collective consciousness and we have lost any fear of it. We are happy to use it as a base for more powerful forms of information management.
What on Earth am I talking about? Sit back down on that chair one more time and I'll tell you. I'm talking about the label system in GMail[4]. Have you seen how that works? There is no great big folder system to speak of. The folders in GMail are created on-the-fly based on the labels you attach to individual e-mails. You are no longer restricted to putting an e-mail in one place and one place only. You are no longer restricted to a hierarchical filing structure. The time honored metaphors of the filing cabinet and the shelf are no longer helpful.
I'm also talking about the faceted classification in WinFS[5]. Think of the way you organize files on a file system. Now, instead of thinking about fixing the names of folders and picking one folder to put your file in, think about that file being, effectively, in as many folders as you like, all at the same time. Last but not least, I'm talking about the emergence of Folksonomies[6] and the intriguing Technorati tagging system[7].
GMail, WinFS, Technorati, all these developments are reticulations of one form or another. It is as if overnight, the world has woken up to the power of it all. (In reality, it is I that has woken up to the power of it all.).
Metadata enthusiasts the world over must be shaking their heads in disbelief. Finally, it's happening - that's great! However, it is not really happening the way the metadata enthusiasts (or theorists) would have predicted. Welcome to the World Wide Web where the only thing that is completely predictable is its total unpredictability.
[1] http://www.apa.org/monitor/mar01/synesthesia.html
[2] http://seanmcgrath.blogspot.com/archives/
[3] http://weblog.hamishharvey.com/mishmash/2005/01/on_hierarchy.html
[4] http://www.beelerspace.com/index.php?p=806
[5] http://msdn.microsoft.com/msdnmag/issues/04/01/WinFS/default.aspx
[6] http://en.wikipedia.org/wiki/Folksonomy
[7] http://www.technorati.com/tag/
Walka trwa (Java vs .NET)
Na sympozjum organizowanym przez TheServerSide paneliści zastanawiali się nad rozwojem Javy. Poodkreślali jest dojrzałość i stabilność, zwrócili uwagę również na witalność platformy .NET. Krytykowano, że .NET jest kontrolowany przez MS (Java zreszta też jest sterowana przez Sun-a). Niesłuszna to krytyka, ponieważ MS uruchomił program CTP (community Technology preview) w którym zaangażował wielu deweloperów w celu nadania kształtu nowym narzedziom .NET. Podkreślono, że Java dla Enterprise powinna być bardziej uproszczona za sprawą nowych szablonów (Framework) np. JavaServer Faces (JFS) lub jej open-source wersja MyFaces (Apache). Na tym polu SFS rywalizuje z Struts (Apache Shale) pod względem funkcjonalności obie technologie się zazębiają (podobnie jest zresztą z EJB 3.0 i JDO jeżeli chodzi o zapamietanie trwałości stanu np. bazy - persistent). Mówiło się, że wersja JSF 2.0 ma mieć funkcjonalność AJAX (asynchronous JavaScript and XML HTTP Request) dla wspomagania tworzenia grubych klientów (tzw. rich Internet application) - chodzi o pobieranie danych z serwera bez przeładowania strony (każde przeładowanie strony oznacza wobec bezstanowej właściwości protokołu HTTP utratę stanu informacji). AJAX będzie rozszerzeniem do przeglądarek Mozilla i Internet Explorer. Czasu jest mało, wiadomo, że JSF 1.2 będzie włączona do J2EE 5 w 2006 roku (i kto tu mówi o opóźnieniu Longhorna), zaś specyfikacja JSF 2.0 będzie zaproponowana wkrótce (bo pod koniec roku) jako JSR (Java Specification Request).
LinuxWorld | Java vs. Microsoft .Net debate rages
Java vs. Microsoft .Net debate rages
Paul Krill, InfoWorld
08/03/2005 08:21:10
After describing Java as a stable technology with no great surprises likely to come, Java experts during a conference panel session fielded questions about the competitive power of Microsoft's rival .Net platform.
Speaking during a session Saturday at TheServerSide Java Symposium, panelists acknowledged the vitality of Microsoft .Net development technology but defended their prized Java platform. The session was entitled, "Future of Enterprise Java Keynote Panel" and featured executives from companies such as Sun Microsystems and BEA Systems.
An audience member identifying himself as a Motorola employee expressed concerns about Java's future. He asked why developers should be confident that J2EE would survive the Microsoft onslaught.
"I can see some of the new (projects) that are in the works. They're all .Net. All the younger developers that I'm associated with are all in .Net," he said.
Panelist Mark Hapner, Web services strategist at Sun, cautioned against relying too much on Microsoft technologies. While .Net is a strong competitor. it is tightly controlled by Microsoft, he stressed.
"Basically, Microsoft sucks the air out of .Net for everything that they classify as being of strong interest to themselves and there really is no place for other contributions," he said.
"If you just do (development) for .Net, you're propping yourself up on (Microsoft's) economic model. They get to change it however they choose," Hapner said.
J2EE, on the other hand, supports a collaborative community, Hapner said. "I think (J2EE is) the place where developers and vendors and open source communities can really work together in a way you can't do in .Net," said Hapner.
A day earlier at the conference, Rod Johnson, founder of the Spring framework, had argued that Java had fended off the .Net challenge. He also served on Saturday's panel.
Another audience member questioned how to lure Microsoft developers into the Java camp and how to simplify Java.
"I think that's where you're seeing a lot of these frameworks develop," such as Hibernate, to make Java easier to use, responded panelist Cliff Schmidt, an official with the open source program office at BEA.
Panelist Dion Almaer, an editor at TheServerSide.com, said he witnessed significant interest in Java at Microsoft's TechEd conference last year. "I was amazed at how interested people were in Java," Almaer said.
JavaServer Faces (JSF) technology is being used to make Java development easier, Hapner said. JSF provides reusable user interface elements for building the visual interface to a Web application. Components are rendered into page elements such as text and hyperlinks.
Citing Microsoft's reputation for providing easy development, Johnson stressed Microsoft's strength in marketing messages about ease of use.
"If Microsoft has something that's easy to use, they have a budget to tell you how easy to use it is," Johnson said.
Commenting after the panel session, Floyd Marinescu, founder of TheServerSide online community for Java developers, said he did not believe there would be many conversions from Java to .Net and vice versa, unless there was a specific need. The platforms do the same thing, said Marinescu. "I think there's always going to be committers," to the different camps, he said. TheSeverSide also has an online site for .Net.
Discussing the future of Java, panelists cited wishes such as boosts for asynchronous messaging. "More of this will come because people realize that there is not (just) a single programming model," in Java, said panelist Gregor Hohpe, enterprise integration practice leader at ThoughtWorks.
But for the bigger picture, the Java stack is maturing, Hohpe said. "I think we're going to see a slow maturation, which is already under way," said Hohpe.
Panelist Linda DeMichiel, EJB 3.0 spec lead and a senior staff engineer at Sun, said she anticipated further simplification of J2EE. Schmidt said he expected open source to play a bigger role in corporate development.
Responding to a question about the viability of service-oriented architectures, panelists responded that many companies already are deploying SOAs without actually referring to it as such. "I think it's actually happening. It's just that some people aren't calling it that," Schmidt said. Users are adopting SOA-related technologies such as asynchronous messaging, he said.
On Friday at the conference, the spec leader for the existing JSF 1.0 specification hailed the technology during a session entitled, "JavaServer Faces: Dead on Arrival or a Raging Success?"
Rebutting criticism about the quality of JSF components, Craig McLanahan, spec lead for JSF 1.0 and a senior staff engineer at Sun, said only the standard components have been at issue. "We're seeing libraries of components (coming from) all over the place," McLanahan said.
Components have been available from companies such as ESRI and IBM, he said. Meanwhile, an alternative implementation of JSF, called Apache MyFaces, is available at Apache, he said.
"I will claim that JSF has been successful for its first year," McLanahan said.
One factor impacting JSF adoption is the large number of applications written for Struts, which presents redundancies with JSF, according to McLanahan, who also is the Struts framework founder. He said he has proposed an Apache project, dubbed Shale, that would get rid of overlaps between the two platforms.
Overlaps would be eliminated in areas such as JSF validation and Struts navigation. Although McLanahan had proposed that Shale become Struts 2.0, it instead has been accepted as a Struts sub-project at Apache.org, he said.
Requirements gathering for the upcoming 2.0 version of JSF is under way, McLanahan said.
Version 2.0 is expected to feature AJAX (Asynchronous JavaScript and XML HTTP Request) functionality, to boost rich Internet application development, said Edward Burns, Sun staff engineer and a spec lead for JSF 1.2. AJAX serves as extensions in the popular Mozilla and Internet Explorer browsers, Burns said after a Birds of a Feather session on Friday.
"It's ubiquitous," he said.
Version 1.2 of JSF, meanwhile, is due as part of J2EE 5 in approximately 2006. It will feature in-progress tree creation and content interweaving. JSF 2.0 will follow some time after that. "We want to get the JSR (Java Specification Request) filed (for JSF 2.0) pretty soon,' by the end of the year, Burns said.
Paul Krill, InfoWorld
08/03/2005 08:21:10
After describing Java as a stable technology with no great surprises likely to come, Java experts during a conference panel session fielded questions about the competitive power of Microsoft's rival .Net platform.
Speaking during a session Saturday at TheServerSide Java Symposium, panelists acknowledged the vitality of Microsoft .Net development technology but defended their prized Java platform. The session was entitled, "Future of Enterprise Java Keynote Panel" and featured executives from companies such as Sun Microsystems and BEA Systems.
An audience member identifying himself as a Motorola employee expressed concerns about Java's future. He asked why developers should be confident that J2EE would survive the Microsoft onslaught.
"I can see some of the new (projects) that are in the works. They're all .Net. All the younger developers that I'm associated with are all in .Net," he said.
Panelist Mark Hapner, Web services strategist at Sun, cautioned against relying too much on Microsoft technologies. While .Net is a strong competitor. it is tightly controlled by Microsoft, he stressed.
"Basically, Microsoft sucks the air out of .Net for everything that they classify as being of strong interest to themselves and there really is no place for other contributions," he said.
"If you just do (development) for .Net, you're propping yourself up on (Microsoft's) economic model. They get to change it however they choose," Hapner said.
J2EE, on the other hand, supports a collaborative community, Hapner said. "I think (J2EE is) the place where developers and vendors and open source communities can really work together in a way you can't do in .Net," said Hapner.
A day earlier at the conference, Rod Johnson, founder of the Spring framework, had argued that Java had fended off the .Net challenge. He also served on Saturday's panel.
Another audience member questioned how to lure Microsoft developers into the Java camp and how to simplify Java.
"I think that's where you're seeing a lot of these frameworks develop," such as Hibernate, to make Java easier to use, responded panelist Cliff Schmidt, an official with the open source program office at BEA.
Panelist Dion Almaer, an editor at TheServerSide.com, said he witnessed significant interest in Java at Microsoft's TechEd conference last year. "I was amazed at how interested people were in Java," Almaer said.
JavaServer Faces (JSF) technology is being used to make Java development easier, Hapner said. JSF provides reusable user interface elements for building the visual interface to a Web application. Components are rendered into page elements such as text and hyperlinks.
Citing Microsoft's reputation for providing easy development, Johnson stressed Microsoft's strength in marketing messages about ease of use.
"If Microsoft has something that's easy to use, they have a budget to tell you how easy to use it is," Johnson said.
Commenting after the panel session, Floyd Marinescu, founder of TheServerSide online community for Java developers, said he did not believe there would be many conversions from Java to .Net and vice versa, unless there was a specific need. The platforms do the same thing, said Marinescu. "I think there's always going to be committers," to the different camps, he said. TheSeverSide also has an online site for .Net.
Discussing the future of Java, panelists cited wishes such as boosts for asynchronous messaging. "More of this will come because people realize that there is not (just) a single programming model," in Java, said panelist Gregor Hohpe, enterprise integration practice leader at ThoughtWorks.
But for the bigger picture, the Java stack is maturing, Hohpe said. "I think we're going to see a slow maturation, which is already under way," said Hohpe.
Panelist Linda DeMichiel, EJB 3.0 spec lead and a senior staff engineer at Sun, said she anticipated further simplification of J2EE. Schmidt said he expected open source to play a bigger role in corporate development.
Responding to a question about the viability of service-oriented architectures, panelists responded that many companies already are deploying SOAs without actually referring to it as such. "I think it's actually happening. It's just that some people aren't calling it that," Schmidt said. Users are adopting SOA-related technologies such as asynchronous messaging, he said.
On Friday at the conference, the spec leader for the existing JSF 1.0 specification hailed the technology during a session entitled, "JavaServer Faces: Dead on Arrival or a Raging Success?"
Rebutting criticism about the quality of JSF components, Craig McLanahan, spec lead for JSF 1.0 and a senior staff engineer at Sun, said only the standard components have been at issue. "We're seeing libraries of components (coming from) all over the place," McLanahan said.
Components have been available from companies such as ESRI and IBM, he said. Meanwhile, an alternative implementation of JSF, called Apache MyFaces, is available at Apache, he said.
"I will claim that JSF has been successful for its first year," McLanahan said.
One factor impacting JSF adoption is the large number of applications written for Struts, which presents redundancies with JSF, according to McLanahan, who also is the Struts framework founder. He said he has proposed an Apache project, dubbed Shale, that would get rid of overlaps between the two platforms.
Overlaps would be eliminated in areas such as JSF validation and Struts navigation. Although McLanahan had proposed that Shale become Struts 2.0, it instead has been accepted as a Struts sub-project at Apache.org, he said.
Requirements gathering for the upcoming 2.0 version of JSF is under way, McLanahan said.
Version 2.0 is expected to feature AJAX (Asynchronous JavaScript and XML HTTP Request) functionality, to boost rich Internet application development, said Edward Burns, Sun staff engineer and a spec lead for JSF 1.2. AJAX serves as extensions in the popular Mozilla and Internet Explorer browsers, Burns said after a Birds of a Feather session on Friday.
"It's ubiquitous," he said.
Version 1.2 of JSF, meanwhile, is due as part of J2EE 5 in approximately 2006. It will feature in-progress tree creation and content interweaving. JSF 2.0 will follow some time after that. "We want to get the JSR (Java Specification Request) filed (for JSF 2.0) pretty soon,' by the end of the year, Burns said.
Jednak będzie patent w Europie
EU Council Passes Software Patent Law
By Nate Mook, BetaNews
March 7, 2005, 1:03 PM
As expected, European governments have approved a controversial bill allowing software patents, despite objections from Denmark, Poland and Portugal. The draft legislation, called the Computer Implemented Inventions Directive, will now be sent back to the European Parliament for a second reading.
The directive's fate is still uncertain, however, as the parliament could demand further changes. The EU Parliament has previously expressed dissatisfaction with the bill, requesting that it be returned for a first reading in February.
The European Commission denied that request for fear of disrupting standard legislative process and setting a precedent for future directives. Internal Market Commissioner Charlie McCreevy says he will work with the parliament and EU governments to ensure a compromise is met.
"I will work to make sure these concerns are taken into account in the interest of a balanced result," McCreevy told reporters.
The software patent bill has met with objection from the start. Polish representatives twice rejected its adoption, while Spain and the Netherlands have supported a restart. In February, hundreds of protesters gathered in Brussels to convince European lawmakers to drop the proposal entirely
By Nate Mook, BetaNews
March 7, 2005, 1:03 PM
As expected, European governments have approved a controversial bill allowing software patents, despite objections from Denmark, Poland and Portugal. The draft legislation, called the Computer Implemented Inventions Directive, will now be sent back to the European Parliament for a second reading.
The directive's fate is still uncertain, however, as the parliament could demand further changes. The EU Parliament has previously expressed dissatisfaction with the bill, requesting that it be returned for a first reading in February.
The European Commission denied that request for fear of disrupting standard legislative process and setting a precedent for future directives. Internal Market Commissioner Charlie McCreevy says he will work with the parliament and EU governments to ensure a compromise is met.
"I will work to make sure these concerns are taken into account in the interest of a balanced result," McCreevy told reporters.
The software patent bill has met with objection from the start. Polish representatives twice rejected its adoption, while Spain and the Netherlands have supported a restart. In February, hundreds of protesters gathered in Brussels to convince European lawmakers to drop the proposal entirely
poniedziałek, marca 07, 2005
Maguma for free (PHP)
Maguma OpenStudio will be the 7th of march 2005, 3 days before CeBIT in Hannover, Germany begins.
view
trackbacks
Submitted by Maguma on Thu, 03/03/2005 - 17:43. Articles Developer
Maguma GmbH will open the source of Maguma Studio, the company's first PHP IDE, naming this distribution Maguma OpenStudio on the 7th of march 2005, 3 days before CeBIT in Hannover, Germany begins.
Maguma OpenStudio will be licensed in Open Source under the Mozilla Public License (MPL) 1.1.Maguma Studio will continue to sell a commercial version of the product for (19,99 Euro) and with the difference to Maguma OpenStudio being the integrated CVS.
Maguma will support Maguma OpenStudio and the community under www.phpwizard.net. Maguma OpenStudio and the documentation will be freely downloadable after the 7 march on www.phpwizard.net, www.maguma.com and www.sourceforge.net.
view
trackbacks
Submitted by Maguma on Thu, 03/03/2005 - 17:43. Articles Developer
Maguma GmbH will open the source of Maguma Studio, the company's first PHP IDE, naming this distribution Maguma OpenStudio on the 7th of march 2005, 3 days before CeBIT in Hannover, Germany begins.
Maguma OpenStudio will be licensed in Open Source under the Mozilla Public License (MPL) 1.1.Maguma Studio will continue to sell a commercial version of the product for (19,99 Euro) and with the difference to Maguma OpenStudio being the integrated CVS.
Maguma will support Maguma OpenStudio and the community under www.phpwizard.net. Maguma OpenStudio and the documentation will be freely downloadable after the 7 march on www.phpwizard.net, www.maguma.com and www.sourceforge.net.
Jak to się robi w Google...
A Glimpse Of Google
By Charles Babcock Courtesy of InformationWeek
Google Inc. owes its success in part to the unique computing architecture it invented to suit its business--serving up searches of the Web to millions of users with lightning-fast speed.
Attendees at the EclipseCon conference in Burlingame, Calif., Wednesday were treated to a glimpse of how that technology was developed in Google's early days and how that technology operates today, courtesy of a speech by Urs Hoelzle, Google's VP of engineering.
EclipseCon is the second annual meeting of users of the Eclipse open-source programmer's workbench, a platform that helps unrelated software-development tools work together.
To invent Google's technology, developers had to throw out assumptions previously used in large data centers and implant new ones, Hoelzle told attendees. And because exactly what will be searched for on any given day is never predictable, keeping the 10 billion pages of the Web close at hand is a daunting challenge.
Hoelzle, a man with a short black beard and quick wit, told audience members that a Google search on "Eclipse" produced more results for solar events and the Mitsubishi car than their favorite development environment. "So everyone here, get back to work," he admonished his listeners, who chuckled in response.
Hoelzle showed pictures of the early Google hardware data center, which consisted of two desktop machines "that no one else was using" in a cluttered setting at Stanford University in 1997. By 1999, it was a large set of thin, rack-mounted Intel servers with a maze of cables coming out the back. By 2000, it was a much cleaner set of 1,000 dual-processor servers in racks that incorporated switching to eliminate the cables.
"The underlying hardware is pretty darn cheap, but achieving scalability has many different aspects," Hoelzle said.
Ensuring reliability was another concern. With so many commodity hardware servers, "expect to lose one a day," he said. Google decided to "try to deal with that in an automated way. Otherwise, you will have lots of people running around trying to restart servers."
Hoelzle then flashed a picture on the screen of six fire trucks at a Google data center. "I can't tell you what happened, but it's not about one machine going down," he said. He didn't disclose when the incident occurred. "No users were harmed in this picture," he added.
To cope with outages of a variable nature, Google built the Google File System, which was closely geared to Google's search computing tasks and had a high tolerance for server failures.
Google operations are built around large files that are broken down into 64-Mbyte chunks and scattered across multiple "chunk servers." A description of each file, its number of chunks, and chunk locations are kept on a master server. Each 64-Mbyte chunk is also replicated on two other servers, so a total of three copies are kept with the path to each retained by the master server.
By scattering its files across many Red Hat Linux servers, Google gains reliability at a low cost. The master server regularly polls chunk servers with a heartbeat message, asking if they're alive. If it fails to get an answer, or if a quick check of the contents on a server indicates that its data has been corrupted, the master server sets about creating a new 64-Mbyte chunk on another server, "usually in a matter of minutes," Hoelzle explained.
Google suffers the loss of a file only if all three copies of the chunk, each on a different server, are lost simultaneously, Hoelzle said. Such a loss would require a lengthy rebuilding process by collecting replacement data off the Web.
Google has indexed the Web through its many Web crawlers that send back summaries of the Web sites they find. Building an index of the Web is a large task that "takes several days on hundreds of machines," he said. The index is renewed constantly.
To search the index quickly, Google breaks it "into pieces called shards," scattered across servers so they may be searched in parallel, each server coming up with part of the answer to a question and feeding it back for aggregated results.
Google's file system, indexing technology, and grid of commodity servers allow it to achieve search times of a quarter of a second on a typical query. The replication and constant heartbeat messaging built into the file system gives it high reliability and availability, he noted.
In addition, as Google servers parse queries, they break them down into smaller tasks and make one trip to the database for a result that may satisfy many users. The process is called "map reduction." Hoelzle said Google once "lost 1,800 of 2,000 map-reduction machines in a large-scale maintenance incident." Because of the load balancing built into the system, Google still completed all queries by steering uncompleted tasks to the machines that showed they had processing power.
"You want to split that huge task into many small tasks, spread across many machines," Hoelzle said. This architecture "makes failure recovery easy. If worker W dies, re-execute the tasks done by that worker elsewhere."
By Charles Babcock Courtesy of InformationWeek
Google Inc. owes its success in part to the unique computing architecture it invented to suit its business--serving up searches of the Web to millions of users with lightning-fast speed.
Attendees at the EclipseCon conference in Burlingame, Calif., Wednesday were treated to a glimpse of how that technology was developed in Google's early days and how that technology operates today, courtesy of a speech by Urs Hoelzle, Google's VP of engineering.
EclipseCon is the second annual meeting of users of the Eclipse open-source programmer's workbench, a platform that helps unrelated software-development tools work together.
To invent Google's technology, developers had to throw out assumptions previously used in large data centers and implant new ones, Hoelzle told attendees. And because exactly what will be searched for on any given day is never predictable, keeping the 10 billion pages of the Web close at hand is a daunting challenge.
Hoelzle, a man with a short black beard and quick wit, told audience members that a Google search on "Eclipse" produced more results for solar events and the Mitsubishi car than their favorite development environment. "So everyone here, get back to work," he admonished his listeners, who chuckled in response.
Hoelzle showed pictures of the early Google hardware data center, which consisted of two desktop machines "that no one else was using" in a cluttered setting at Stanford University in 1997. By 1999, it was a large set of thin, rack-mounted Intel servers with a maze of cables coming out the back. By 2000, it was a much cleaner set of 1,000 dual-processor servers in racks that incorporated switching to eliminate the cables.
"The underlying hardware is pretty darn cheap, but achieving scalability has many different aspects," Hoelzle said.
Ensuring reliability was another concern. With so many commodity hardware servers, "expect to lose one a day," he said. Google decided to "try to deal with that in an automated way. Otherwise, you will have lots of people running around trying to restart servers."
Hoelzle then flashed a picture on the screen of six fire trucks at a Google data center. "I can't tell you what happened, but it's not about one machine going down," he said. He didn't disclose when the incident occurred. "No users were harmed in this picture," he added.
To cope with outages of a variable nature, Google built the Google File System, which was closely geared to Google's search computing tasks and had a high tolerance for server failures.
Google operations are built around large files that are broken down into 64-Mbyte chunks and scattered across multiple "chunk servers." A description of each file, its number of chunks, and chunk locations are kept on a master server. Each 64-Mbyte chunk is also replicated on two other servers, so a total of three copies are kept with the path to each retained by the master server.
By scattering its files across many Red Hat Linux servers, Google gains reliability at a low cost. The master server regularly polls chunk servers with a heartbeat message, asking if they're alive. If it fails to get an answer, or if a quick check of the contents on a server indicates that its data has been corrupted, the master server sets about creating a new 64-Mbyte chunk on another server, "usually in a matter of minutes," Hoelzle explained.
Google suffers the loss of a file only if all three copies of the chunk, each on a different server, are lost simultaneously, Hoelzle said. Such a loss would require a lengthy rebuilding process by collecting replacement data off the Web.
Google has indexed the Web through its many Web crawlers that send back summaries of the Web sites they find. Building an index of the Web is a large task that "takes several days on hundreds of machines," he said. The index is renewed constantly.
To search the index quickly, Google breaks it "into pieces called shards," scattered across servers so they may be searched in parallel, each server coming up with part of the answer to a question and feeding it back for aggregated results.
Google's file system, indexing technology, and grid of commodity servers allow it to achieve search times of a quarter of a second on a typical query. The replication and constant heartbeat messaging built into the file system gives it high reliability and availability, he noted.
In addition, as Google servers parse queries, they break them down into smaller tasks and make one trip to the database for a result that may satisfy many users. The process is called "map reduction." Hoelzle said Google once "lost 1,800 of 2,000 map-reduction machines in a large-scale maintenance incident." Because of the load balancing built into the system, Google still completed all queries by steering uncompleted tasks to the machines that showed they had processing power.
"You want to split that huge task into many small tasks, spread across many machines," Hoelzle said. This architecture "makes failure recovery easy. If worker W dies, re-execute the tasks done by that worker elsewhere."
PC World | Internet: Visual appeal
Site of the month
Getting the right graphics for your pages often makes or breaks a site. And for this column, there's not so much one site of the month as several that come highly recommended.
While sites such as www.freeimages.co.uk provide some useful graphics, top-quality photography usually requires some payment. With that in mind, I suggest visiting the big-name stock image agencies like Getty Images (www.gettyimages.com), Corbis (www.corbis.com) and Image 100 (www.image100.com). Getty and Image 100 offer some graphics royalty-free for personal use
Getting the right graphics for your pages often makes or breaks a site. And for this column, there's not so much one site of the month as several that come highly recommended.
While sites such as www.freeimages.co.uk provide some useful graphics, top-quality photography usually requires some payment. With that in mind, I suggest visiting the big-name stock image agencies like Getty Images (www.gettyimages.com), Corbis (www.corbis.com) and Image 100 (www.image100.com). Getty and Image 100 offer some graphics royalty-free for personal use
Osiołkowi w żłoby dano...(SVG vs. Flash)
SVG And Flash As The Same Species
We can name several technologies for displaying 2D interactive vector graphics on the web however, two names rises among other contenders.
These are the Macromedia's Shock Wave Flash (SWF) and the World Wide Web Consortium's (W3C) Scalable Vector Graphics (SVG). Unlike Flash SVG is not for sale.
Adobe's SVG browser plug-in has achieved barely around 10% penetration, compared to Macromedia's 90% for the Flash plug-in. Flash has already established an outstanding standard nevertheless, this did not make SVG an inferior technology.
In the past Flash was considered a de-facto standard under control of one single vendor. For this reason, W3C proposed a recommendation to achieve standardized 2D interactive vector graphics on the web. This resulted to the birth of SVG. Its first version was Recommendation 1.0 which was introduced in 2001. The current version though is 1.1 which was released on January 2003.
Some people asserts that SVG is as powerful as Flash. Some say it is superior. But still there are others who think it's the other way around. Which is which really?! To help you decide here are some vital facts:
Use Flash in the following instances:
1. You must use Flash if you want to make a Flash-like website. To replicate it using SVG is hard to do.
2. If you want to create complex animations or games you can use any of the two. However, you must be aware that SVG's built-in SMIL animation engine is very processor exhaustive. Conversely, good results can be achieved using JavaScript animations.
3. If your users are not that computer literate like for example children for a children's site or a site appealing to a broad audience, you must use Flash.
4. SVG supports sound. Nonetheless, if sound involves important intricate details, use Flash.
5. Use Flash if you prefer WYSIWYG to script.
SVG is advantageous in the following instances:
1. Use SVG if your task is fully scriptable. You can try DOM1 (part DOM2) interface and JavaScript. It means that you can build your empty SVG image using JavaScript.
2. SVG can effortlessly be created by ASP, PHP, Perl and extracted from a database. Be cautious though in setting mime-types on the server.
3. It is XML meaning it can be read by anything that can read XML. Flash can use XML but it has to be converted first.
4. You need not apply Flash's action-script because SVG does not code per browser. It has a built-in ECMA-script engine.
5. SVG runs on IE/NS4+NS6/Mozilla and on PC/MAC/Linux.
6. SVG can be transformed through an XSLT style-sheet or parser. It also supports standard CSS1 style-sheets.
7. Texts used remains selectable and searchable. Graphic "objects" can be grouped, styled, transformed or composited and embedded text can be searched or indexed.
8. Features of the specification include nested transformations, clipping paths, alpha masks, filter effects, template objects and, of course, extensibility
9. You need not buy Flash since only texteditor is needed to create SVG.
By now, I know you already have a choice. Take your pick and enjoy the features of your chosen technology
JDO 2, too little, too late
Nowa (2-ga) specyfikacja JDO to tylko draft (4-8 miesiecy potrwa proces approval), skrytykowana przez uczestników JCP poniewaz specyfikacja EJB 3.0 znacznie ja przewyzsza funkcjonalnie jezeli chodzi o zapewnienie trwalosci (persistant) - stad powyzszy tytul.
Java object spec set for open source Apache licensing
Paul Krill, InfoWorld
07/03/2005 08:02:16
Java Data Objects (JDO) 2.0, a specification for Java object persistence that has been controversial of late, is to be offered under Apache open source licensing, said a Sun Microsystems architect at the TheServerSide Java Symposium event on Friday.
The more liberal Apache licensing will allow users to do what they wish with the technology, said Sun architect Craig Russell. Version 1.0 of JDO had been offered under the more-restrictive Sun Community Source License, Russell said.
JDO 2.0 features improvements in areas such as query functionality.
JDO was recently the subject of a petition drive after an initial vote within the JCP (Java Community Process) rejected the specification. A new vote announced this week approved a public draft. But JDO 2.0 still must be subjected to a four- to eight-month final approval process by the JCP Executive Committee, Russell said.
The EJB 3.0 specification is expected to supersede JDO as a method of Java persistence, although Russell noted users can still deploy JDO.
"JDO will live on as an Apache project, no question about it," Russell said.
"The future evolution of JDO will depend on the needs of the Java community," Russell added.Â
The petition drive last month in support of JDO 2.0 drew more than 1,000 signatures after the JCP's initial tally on approval fell short, 10 votes to five. JCP member JBoss, in voting no, said JDO 2.0 constituted more than a maintenance release and would cause confusion with EJB 3.0.
Also at the TheServerSide event, an Oracle executive said the company plans to upgrade its portal-development technology by adding the ability to do portal development via the company's JDeveloper development tool. Currently, development is browser-based.
"(The improvements) will allow a more rich design time because you can do more in a desktop tool (like JDeveloper) than a browser," said the official, Ted Farrell, chief architect for the Application Development Tools Division at Oracle.
The new functionality is enabled via support of JavaServer Faces technology for building Java-based Web applications.
Java object spec set for open source Apache licensing
Paul Krill, InfoWorld
07/03/2005 08:02:16
Java Data Objects (JDO) 2.0, a specification for Java object persistence that has been controversial of late, is to be offered under Apache open source licensing, said a Sun Microsystems architect at the TheServerSide Java Symposium event on Friday.
The more liberal Apache licensing will allow users to do what they wish with the technology, said Sun architect Craig Russell. Version 1.0 of JDO had been offered under the more-restrictive Sun Community Source License, Russell said.
JDO 2.0 features improvements in areas such as query functionality.
JDO was recently the subject of a petition drive after an initial vote within the JCP (Java Community Process) rejected the specification. A new vote announced this week approved a public draft. But JDO 2.0 still must be subjected to a four- to eight-month final approval process by the JCP Executive Committee, Russell said.
The EJB 3.0 specification is expected to supersede JDO as a method of Java persistence, although Russell noted users can still deploy JDO.
"JDO will live on as an Apache project, no question about it," Russell said.
"The future evolution of JDO will depend on the needs of the Java community," Russell added.Â
The petition drive last month in support of JDO 2.0 drew more than 1,000 signatures after the JCP's initial tally on approval fell short, 10 votes to five. JCP member JBoss, in voting no, said JDO 2.0 constituted more than a maintenance release and would cause confusion with EJB 3.0.
Also at the TheServerSide event, an Oracle executive said the company plans to upgrade its portal-development technology by adding the ability to do portal development via the company's JDeveloper development tool. Currently, development is browser-based.
"(The improvements) will allow a more rich design time because you can do more in a desktop tool (like JDeveloper) than a browser," said the official, Ted Farrell, chief architect for the Application Development Tools Division at Oracle.
The new functionality is enabled via support of JavaServer Faces technology for building Java-based Web applications.
Subskrybuj:
Posty (Atom)