Yahoo! Releases OSS Ajax and Design Tools

Cocteaustin writes "Today Yahoo! released the Yahoo! User Interface Library. This library is comprised of a number of dynamic HTML utilities and controls for building rich web UIs and Ajax applications. They are made available under an open-source license. In addition, Yahoo! released the Yahoo! Design Pattern Library. This collection of design patterns for Web interaction is intended to provide Web designers prescriptive guidance to help solve common design problems on the Web. Both are free in both senses of the word."

Comments

I for one find that...

[gd23ka]

awesomely generous. Hope there are no patent string attached.

Re:I for one find that...

[zenquest]

BSD License: http://developer.yahoo.net/yui/license.txt

Re:I for one find that...

[AKAImBatman]

I seriously doubt that there are any real patents or other strings attached. I think this is more of a publicity and goodwill stunt more than anything else. As far as I can tell, there are no sophisticated components here, just the basic stuff that most AJAX developers already have in their toolkits.

The list of components is:

* Calendar
* Slider
* TreeView

That's a pretty small list, and all are components that are fairly common in AJAX circles.

The core utilities portion of the library is just Yahoo's convenience methods that help abstract away browser differences. Nice if you don't have wrappers like these already, but not very useful if you do. Many AJAX programmers will probably choose to stick with their own libraries.

A few things that come to mind that are missing from this library are:

* A text editor components
* DataGrid/Spreadsheet component
* Scrolling viewports
* Feature-rich DHTML replacements for buttons, lists, radio buttons, and other common controls.
* Application layout engine

I'm pretty sure that Yahoo! has these types of components, but isn't going to share as long as there is more value in keeping them secret.

All in all, it's a nice gesture by Yahoo!. Just don't expect a complete library. :-)

Re:I for one find that...

[Anonymous Coward]

Actually some of the components are very sophisticated indeed. The drag and drop library is the best I've seen - it can handly logical groups of related elements, point intersection or area overlay modes, "padding" for drag detection areas, configurable drag triggers settings and more.

Re:I for one find that...

[AKAImBatman]

Like I said, if you don't have it in your library already, it's a nice thing to have. If you do, then it's probably so much noise.

Re:I for one find that...

[fean]

This library probably wasn't released for the benefit of those already waist-deep in AJAX / Javascript-UI-Niceness... I'm guessing if you already have a very complex drag-and-drop piece of code in your library, nothing that was released today would be any help...

That said, I think it's positively wonderful! The hard part of web design isn't usually the design, but convincing the clients that a certain technology is OK to use. It's even harder when you're working in Microsoft land, and everyone knows you're a huge firefox supporter (I get several emails a day from people who surf in firefox for the express reason of finding sites that render incorrectly, so they can point out that not all sites look good in firefox... there is no reasoning with them, just deleting their emails)

Having Yahoo! release these to the wild will make it much easier to use this functionality, just by saying "How about we use the Yahoo UI Widget here?"

Re:I for one find that...

[Anonymous Coward]

Oh any patent issues will come up 3 years from now once everyone's using it.

Re:I for one find that...

[Anonymous Coward]

I've tried a whole bunch of drag-n-drop libraries in the past and the Yahoo! one is the best I've seen. This stuff isn't aimed at newbies - it's aimed at developers working on some of the most high traffic sites in the world.

Re:I for one find that...

[papukanghi]

I seriously doubt that you got the point. The whole point of this library is not the Widgets. The library itself offers you ways to write your own widgets.

* A standard way to do Ajax.
* A way to have your own events.
* A way to do positioning. (moving objects)
* A way to do animation.
* All that in one - three lines of code.

Include the library and call the methods. Simple. How difficult would it be to implement a graphical color picker be wheh you have the slider already implemented for you? And if its well written and you make it avaliable as BSD licence why will yahoo/others not use that rather then writing their own?

That is the point of releasing the libraries. So that the best solutions become the standard. Its not nessesiraly a one way (Yahoo -> us) process.

Re:I for one find that...

[AKAImBatman]

I seriously doubt that you got the point.

You don't write much AJAX/DHTML, do you?

* A standard way to do Ajax.

More standard than the DOM? There is no "standard" here or there. There's an API and suggested design methods. That and 50 cents will buy you a cup of coffee.

* A way to have your own events.

Your own events? You mean, pass the function handle to a given object, then wait for it to call you back with an event object? This is basic JavaScript.

* A way to do positioning. (moving objects)

It's called CSS. It's part of the DOM. Depending on what type of graphics you're doing, it usually only takes a few lines of code to create an object with an API that wraps these data elements.

* A way to do animation.

You definitely don't do much AJAX, do you? The only reason why it hasn't replaced Flash is because there's no reliable method for rotating and shearing graphics. Irregular shapes also represent a serious problem for DHTML, but one that can be worked around. Once SVG is standard in every browser, AJAX will be finally capable of replacing Flash for web animations.

* All that in one - three lines of code.

Correction, all that in ~10,000 lines of code from Yahoo!, plus your 3 lines of code for a simple demo. By the time you're done doing what you need to do, it will be ~20,000 lines of code from you, ~10,000 form Yahoo!, and only ~3,000 lines of Yahoo!'s actually utilized.

I've said it before, and I'll say it again: Most of us already have this stuff. If you yourself don't, then have at it. It's a nice library. It's just not Earth shattering.

How difficult would it be to implement a graphical color picker be wheh you have the slider already implemented for you?

Hard. You'd be amazed at how many niggling details are inherent in trying to write a good component. Which is my point. We already all have the "core" utilities. For most AJAX programmers, that's not a useful thing. What we don't have are well-designed, well-debugged, and well-tested components. Yahoo! shows an example of a color chooser on the slider page. They conveniently forget to mention, however, the thousand or so lines of code that it took to implement that. I would have much rather they taken that Color Chooser, make it into a pluggable component, give it a proper API and feature set, and then test the hell out of it. That would have been far more useful to me than a slider that I already have with a demo of a non-reusable component.

Re:I for one find that...

[psionic]

hmmm I wonder if viewers of pages using this "Free" code will be forced to install the Yahoo! Toolbar to view the content :-)

Yahoo is the new Google?

[Teetow]

So, while Google is expanding its new evil empire, Yahoo is courting indie developers? Strange days on planet earth...

Re:Yahoo is the new Google?

[hsmith]

more like, they are both competing for us to use their products. no matter what, we don't lose in the end. as long as the tools are free, i don't see a problem

Re:Yahoo is the new Google?

[ELProphet]

Truly strange. The first product from Yahoo I have ever wished Google came out with first... earily freaky...

Re:Yahoo is the new Google?

[m50d]

Yahoo has always been like this, it's just people didn't notice while google was the new hotness. Seriously, they seem to be doing the Right Thing, and it's about time they got some recognition.

Re:Yahoo is the new Google?

[stupidfoo]

Very true. Their developer APIs are the best of any major offering.

Check them out here

Their stated goal is to have startups use their APIs as the foundation for new sites/tech.

Re:Yahoo is the new Google?

[jallen02]

Never mind that they (Y!) censor searches in China just like google. Google gets flamed for it while all of the other major search engines in China do the same thing so they don't get banned.

Jeremy

Re:Yahoo is the new Google?

[Anonymous Coward]

"The Right Thing?" You mean like this? ... http://www.rsf.org/article.php3?id_article=16402

Re:Yahoo is the new Google?

[m50d]

Erm, BS. Yahoo got nothing but flames when the story about them doing it was posted. Google got 3/4 of people saying "they have no choice, it's better than not being accessible from China at all". Google gets as much love as Apple, and it just kills me when people try and claim they're being persecuted.

Re:Yahoo is the new Google?

[Anonymous Coward]

"they seem to be doing the Right Thing"

Except maybe when they help China to jail pro-democracy dissidents...

Re:Yahoo is the new Google?

[anomalous cohort]

Their developer APIs are the best of any major offering.

That was interesting. Thanks for the link. It looks like yahoo is using JSON for their wire protocol. I find that to be interesting for two reasons. The first is that it is not an XML based protocol. The second reason is that JSON is also the wire protocol for the ironically named AJAX.NET.

I wonder who else is developing JSON based APIs and if that is going to be the next big thing?

Re:Yahoo is the new Google?

[jallen02]

They should all get flamed equally. Anyone that plays those games. I don't really put that much stock in Google or any web based company like that. Search is great. Having everything I want to be private on a Google server.. no thanks :)

J

Re:Yahoo is the new Google?

[Anonymous Coward]

> I wonder who else is developing JSON based APIs and if that is going to be the next big thing?

Lots of toolkits use JSON. Mochikit. DWR. Dojo. JSON's not bad, but it's nice to have a toolkit that supports multiple transports. SOAP may be a dead duck (or at least it should be) but it's nice to be able to sling arbitrary XML around with the same API's as you would use for JSON. To say nothing of the venerable old POST format for legacy sites: dropping in dojo with proxomitron blows the doors off of greasemonkey functionality-wise, but without an easy POST api, it's pointless for many apps.

Lord save me from "the next big thing". Once the W3C gets its hands on it, it's no longer useful as a standard.

Re:Yahoo is the new Google?

[JacksBrokenCode]

Yahoo has always been like this, it's just people didn't notice while google was the new hotness

Or perhaps people were pre-occupied with Yahoo!'s problem of being linked to spyware funding. People are worried about Google's response to the government subpoena for info, but seem to turn a blind eye to Yahoo! complicity. People are blasting Google for censoring Chinese search results, but the Shi Tao PR flap seems to have blown over for Yahoo!.

Or are you talking strictly in the context of free code for developers? There have been complaints that Google uses OSS but doesn't contribute enough back, yet their overal social & business ethics seem to be better than Yahoo!'s. Does Yahoo!'s code-release to the developer community somehow mitigate their other poor ethics? And how much redemption is there in a codebase that many developers have already created on their own?

Re:Yahoo is the new Google?

[Anonymous Coward]

User "JacksBrokenCode" wrote:
And how much redemption is there in a codebase that many developers have already created on their own?

Well, Jack, how about "it's not broken"?

Re:Yahoo is the new Google?

[m50d]

Or perhaps people were pre-occupied with Yahoo!'s problem of being linked to spyware funding.

Hahahahaha. And google does precisely the same thing, as you'd see if you read a couple of paragraphs into the article you link to.

People are worried about Google's response to the government subpoena for info, but seem to turn a blind eye to Yahoo! complicity.

Worried? I see more mindless admiration "OMG they're so awesome for trying to fight this in court". And IIRC Yahoo made it clear everything was anonymised, something that Google's been very quiet about.

People are blasting Google for censoring Chinese search results, but the Shi Tao PR flap seems to have blown over for Yahoo!.

Look back at that threads, 3/4 of the posts in the google one were saying "They have no choice, if they want to operate in China they have to follow their laws". Yahoo got nothing but flames.

There have been complaints that Google uses OSS but doesn't contribute enough back, yet their overal social & business ethics seem to be better than Yahoo!'s.

How so? Remember when google takedown'd the guy offering RSS feeds of google news?

And how much redemption is there in a codebase that many developers have already created on their own?

Plenty, just look at all the database vendors getting it. But that's beside the point.

Really good stuff

[Vivek Jishtu]

I tried out some of the JavaScript code they are offering. It is a nice library of functions for web application development.

Re:Really good stuff

[diegocgteleline.es]

"I tried out some of the JavaScript code they are offering. It is a nice library of functions for web application development."

Really good stuff (Score:3, Informative)

Oh, well. Slashdot is based in what people thinks it's informative or not, but if people can't tell if something is informative or not, then moderation isn't useful.

Design Fixes.

[Anonymous Coward]

"Patterns are optimal solutions to common problems. As common problems are tossed around a community and are resolved, common solutions often spontaneously emerge. Eventually, the best of these rise above the din and self-identify and become refined until they reach the status of a Design Pattern."

According to some. Some languages don't need "design patterns" because they don't have the problems the pattern is ment to fix.

Re:Design Fixes.

[Anonymous Coward]

Some languages don't need "design patterns" because they don't have the problems the pattern is ment to fix.

That's like saying "NP-Complete" problems are easy in some languages.

Re:Design Fixes.

[aug24]

Total Arse.

Patterns are nothing to do with languages. Patterns are not meant to fix problems in languages, they are conceptual repeating patterns, like 'the need to store', 'the need to display', 'the need to pass data'.

If your language of choice happens to implement one of these languages (roughly like struct or Object for the DTO pattern), then so much the better.

Justin.

Re:Design Fixes.

[ajwitte]

What language are you thinking of? Prolog or some other non-imperative language? I don't know Prolog, but I'm sure there are "design patterns" - time-tested ways of satisfying certain needs - even if they aren't the same as the ones use with imperative languages.

Re:Design Fixes.

[WarwickRyan]

Mod parent up.

RecordSet's another extremely common pattern which is implemented in most modern languages.

Re:Design Fixes.

[samuel4242]

NP-complete problems are pretty easy (i.e. linear) when you encode them in unary. (Base 1). The size of the encodings become exponetially large so the computation time to solve the problem is now linearly related to input size. It's all a question of finding the right language.

Re:Design Fixes.

[Anonymous Coward]

I agree with you in The Real World, but another important thing to keep in mind is that the neccesity of a design pattern points out somewhere that your development environment diverges from some theoretical ideal perfect programming environment (I say environment not language, because really libraries are where most of this ends up being fixed).

There is a theoretical perfect programming environment in which it is not neccesary to know any design patterns, because they are all "taken care of" and implicit in your available tools. Or in other words, all common design patterns have been factored into your available language/libraries/modules.

This language will probably never exist, but it doesn't hurt to think in that frame of reference in an attempt to improve your programming environment.

show me the money

[Douglas Simmons]

With google's toys they all have mass appeal and drive traffic to the site, ultimately helping google's brokerage. This, while nice for some of us, doesn't. Why would Yahoo bother?

Re:show me the money

[generic-man]

Because they want to improve their image in the open source community, making people think better of Yahoo! when it comes time to choose between Yahoo!, Google, and Brand X for their next enterprise service purchase. I also imagine that they could release code in the future that makes it easy to incorporate Yahoo!'s ad technology so that Web 2.0 developers can contextually-advertise and make money from their efforts.

Google's acts of "driving people to its site" do nothing for Google's bottom line. Google, like Yahoo!, is an advertising company which makes the vast majority of its income from other web sites besides their search engines / portals.

Re:show me the money

[Ilgaz]

Large portion of FreeBSD and Opensource respect from industry (and end users) came from Yahoo.

When people questioned seriousness of that OS, you could (and still) say "Yahoo runs on it". Conversation is over. :)

I have no idea why Google is "good guy" and Yahoo gets amazing misinformed comments on each story. They even called Yahoo "wannabe" when they advertised (existing for years) http://search.yahoo.com/

(robots.txt exclusion, it exists at least since 1999 if you look to archive.org)

Re:show me the money

[aztektum]

"Google's acts of "driving people to its site" do nothing for Google's bottom line. Google, like Yahoo!, is an advertising company..."

You contradicted yourself. Driving people to their site is how Google generates advertising revenue (maybe not all of it, but enough to impact thier "bottom line"). You didn't think all those ads on the right of your search results were free, did you?

Re:show me the money

[Cocteaustin]

This isn't just a "publicity stunt" as some have asserted, far from it. Some of these libraries and patterns have been around in one form or another literally for years within Yahoo!. The people who drove their release as open source were not PR or marketing people but the engineers themselves. The hope, as with any release of open source code, is that by engaging with the community we will get a better outcome for users -- that's ultimately how the business succeeds. Simple as that.

Re:show me the money

[generic-man]

Google makes a ton of money from AdSense ads on other web pages as well. According to their 2004 annual report (warning: tiny fonts and huge amounts of text) page 26, in 2004 Google made 48% of its overall revenues from web sites outside its own domain ("Google Network web sites") versus 52% of its overall revenues from AdWords on its own domain. OK, I exaggerated when I said it "does nothing for Google's bottom line," but advertising on Google's own properties might not grow as much when you compare their sites to the rest of the AdSense-using Internet.

Re:show me the money

[Baricom]

You Yahoos don't get nearly enough credit for what you do (and I have no idea why Google's reputation is invincible - I've seen more than enough for them to be on my "evil" list).

Congratulations on your release, and thank you for continuing to do amazingly cool stuff. Oh, and make sure you continue to keep your marketing droids in check, mmkay? ;)

Under BSD too...

[Anonymous Coward]

That's pretty incredible. Is this out of character, or did the folks at Yahoo! get sick of poorly implemented AJAX sites?

Very nice - great little library

[us7892]

Both pages are clear and the library actually looks very good. Usually, Yahoo is playing catch up to Google, or so it has seemed. This time, Yahoo gets the upper hand. Google is becoming Yahoo, and Yahoo is becoming what Google used to be. Good stuff!

Not that any of this is ground-breaking, but it is a nice little package.

Makes Google's download package from last month look pretty lame.

Re:Very nice - great little library

[ripcrd]

So does yahoo now have "hand"?

Re:Very nice - great little library

[bryguy5]

Both Google and Yahoo have been getting a lot of bad press lately so they need to do something. It's good to see yahoo follow Googles "high ground" even if Google's straying a bit. This is much better PR than re-running analysis on turning people over to oppressive foreign governments (Yahoo) or collaborating with governments to cover up and control information and make sure the truthspeech gets out (Google).

It's goingt to be a long road for both of them as growing resentment http://www.useit.com/alertbox/search_engines.html to their status as gatekeepers builds up.

Blatant plug! Death to the gatekeepers!!
Step 2 ??? Step 3 $$$$ - Forget Tech, you need candy http://www.ilovefundraising.com/

Re:Very nice - great little library

[podperson]

Google also has some catch-up work to do on getting Chinese dissidents imprisoned.

OK OK mod me flamebait.

Re:Very nice - great little library

[B0red At W0rk]

Wow, nice PR move sweeps away search engine fanboys everywhere. /cynicism

Re:Very nice - great little library

[Anonymous Coward]

Regarding UI, yahoo went far beyond google with their new yahoo mail (beta). I didn't know a web browser could do all this kind of stuff.

Re:Very nice - great little library

[Anonymous Coward]

Google's download package from last month made itself look pretty lame

Yahoo UIL and Google Code pages

[digitaldc]

The Yahoo UIL page and the Google Code pages are both useful and coincidentally look quite similar.

Re:Yahoo UIL and Google Code pages

[Anonymous Coward]

Looks more like Apple's dev sites to me.

BSD license

[Rob T Firefly]

Yahoo are releasing this stuff under the BSD License.

Re:BSD license

[Petronius]

yet the menu bar on the right shows a Creative Commons attribution.

Re:BSD license

[Otter]

It's a trick to avoid complaints from Richard Stallman -- this way he won't know what to denounce first, and he'll simply emit smoke.

Re:BSD license

[Ilgaz]

One of the reasons could be that Yahoo uses and advertises FreeBSD since they started.

I know at least 1 giant company started using FreeBSD because of Yahoo. "It can handle entire yahoo userbase" is really a huge "selling" point.

http://www.ictp.trieste.it/~cfonda/sudan/OSs/refer ences/freeBSD/Yahoo_and_FreeBSD.html

Re:BSD license

[Cocteaustin]

Two separate things were released: a Javascript library under BSD, and a design pattern guide under Creative Commons.

Looks like asp.net Atlas stuff

[figleaf]

This looks like asp.net Atlas might be worth investigating though.

When is a design pattern not a Design Pattern?

[BarryNorton]

When it's a UI idiom...

wow

[Abstract_Me]

And they even used teh BSD license.... im speachless...

Hearts and minds

[Anonymous Coward]

This, while nice for some of us, doesn't. Why would Yahoo bother?

Simple, hearts and minds. Right now Google's #1 asset has nothing to do with their search engine, it has everything to do with the fact that they have (or had judging by recent stock hits) the hearts and minds of the public. It's all about the cult of personality that has grown around Google that allows it push as it has. For Yahoo to continue to be successful, they need to gain footing in this area. For them, releasing this can help take some of the steam out of the nerd love fest that is Google (ok, not a lot, but planting the seeds).

Nice Accessibility

[aliens]

If you look through their Design Patterns you'll see that each has an Accessibility section. Very nice addition and often over looked.

Re:Nice Accessibility

[Anonymous Coward]

A nice idea but not a nice addition in this case. It looks to me like appeasement rather than commitment. Most of the accessibility sections simply state that the user should be able to tab to a link/tab/whatever and use the enter key to activate it, i.e. don't break the default UI.

The more insidious examples suggest breaking the UI horribly. For example Auto Complete:

"Allow the user to complete the form by pressing the Tab or Enter keys."

Some, like Drag and Drop, omit the accessibility section altogether. In these cases I guess it was too hard to offer equivalent access for all users, so it's just been ignored rather than flagged as potential problem.

It's cool that they are releasing this stuff, but let's not get carried away by suggesting that what they've offered in these notes gets close to accounting for the accessibility issues raised.

Re:Nice Accessibility

[wk633]

Very good point. There's absolutely no reason to not provide a keyboard interface to re-order modules. Drag and drop makes no sense to a blind user, but re-ordering the linear content sure does. Unfortunately, sometimes it takes a lawsuit to get anything done in this country. http://news.com.com/Blind+patrons+sue+Target+for+s ite+inaccessibility/2100-1030_3-6038123.html/

thanks yahoo!

[DeveloperAdvantage]

This is good to see. Reuse for GUI widgets is one area where reuse has had an impact, in fact, GUI widgets are the most commonly reused components. There are a number of vendors for things like VB controls.

The components and patterns Yahoo has released will speed up the development of feature rich sites for other organizations.

Prototype still rocks

[esconsult1]

After using prototype.js for a while now, its hard to switch to a fatter library which is what the Yahoo library seems like. Each one has their good points, and pieces missing, but I think if you decide to use either, you can't go wrong.

There are some good snippets in there though, and Yahoo has done a good job of introducing code and web services to the developer community, much much more that Google has.

The design patterns are a very very good thing to expose. Although many of us might have been using similar standards, it sort of brings a number of them under one umbrella and into one place.

Re:Prototype still rocks

[at_18]

Prototype may rock, but the website (http://prototype.conio.net) sucks. It's only a page with a download link. So WTF is prototype? Where's the manual, or at least a quick overview of what it does? Not even the .tar.gz file with the library has anything resembling a function list.

I had to google around to find documentation, such as this site).

Re:Prototype still rocks

[lukewarmfusion]

My biggest complaint about prototype is the lack of clear documentation. I have the same problem with script.aculo.us, which has a wiki that is often useless to me.

For a really lightweight effects library, check out moo.fx.

Re:Prototype still rocks

[mythz]

Umm,

Where is the tree widget for Prototype again? OMG there is none!

Prototype has some nice features and allows for OO like programming in Javascript. But it mangles the Javascript Object and the OO design promotes code bloat.

Anyone doing any serious DHTML should check out JQuery (http://jquery.com/), it's a revolutionary lightweight Javascript library that borrows the niceties of Prototype but allows you to achive the same result in a fraction of the code. You can also use it with Prototype, and it weighs in at 10k to boot!

Re:Prototype still rocks

[Anonymous Coward]

I'll admit that the library sucks for documentation but third party docs on both the script.aculo.us site and on http://www.sergiopereira.com/articles/prototype.js .html both have decent docs for it they aren't easy to find but they are there and actually pretty good once you find the core ones ;).

Re:Prototype still rocks

[MemoryDragon]

Prototype has some serious issues, it messes around with basic javascript datastructures, has no namespaces, reserves often used names and keywords $ for instance for itself Many people switched from prototype to dojo exactly for those reasons.

Hurrah! Clap-clap!

[SilentOneNCW]

I say, good show! As early comments have already noted it is indeed a strange month - in which Google is falling, losing popularity due to their stock prices and the whole China debacle, while Yahoo is rising, supporting OSS with a suprisingly useful package. I wonder if this is merely another bump in Google's ultimate victory or a shift in the paradigm, a potentially fatal one for Google. However, let this not take away from the original point of the article -- Congratulations, Yahoo!, and thank you. Your generosity will be remembered.

Re:Hurrah! Clap-clap!

[Anonymous Coward]

Google is falling, losing popularity due to their stock prices and the whole China debacle ... Congratulations, Yahoo!, and thank you. Your generosity will be remembered.

Sure, Google's trapped in a "China debacle," while Yahoo! is only helping the Chinese government throw pro-democracy dissidents in prison.

But, hell, release a couple of Javascript files and we'll forgive 'em for tracking their users' behavior and giving it away to totalitarian dictatorships. Clap motherfucking clap indeed.

"Library", are you kidding me?

[rtilghman]

This is a collection of, count em, THREE main scripts folks. There are free libraries of javascript code out there with orders of magnitude more DHTML functions and scripts. Sure, Yahoo offers some derivatives of each of their primary functions, but one of the categories is a collection of "vented menuing" scripts that could have been written five years ago. Only a multi-national company bent on branding (and yes Google, you're in the same boad) could put up a page like that and call it a Library.

To be honest, I'm consistently frustrated by the status of OSS code with regard to the DHTML components necessary to support open source RIA technology. If you want to do a vented menu, have a slider control, or YADDA you can find about 450 million scripts scattered across the javascript repositories of the web.

What it comes down to is this; if you want to do a collapsible menu or drag and drop then you're in luck, we have the widgets in OSS for you! OSS RIA won't be feasible until SVG stabilizes and is as ubiquitous as the Flash plug-in.

-rt

Re:"Library", are you kidding me?

[AKAImBatman]

"Library", are you kidding me?

Indeed. Most of the posters obviously didn't do much investigation, or are not that familiar with AJAX development. This is the same stuff you've been able to get elsewhere for a LONG time. The Blueshoes and ActiveWidget collections are a lot more useful, albeit not entirely free.

To be honest, I'm consistently frustrated by the status of OSS code with regard to the DHTML components necessary to support open source RIA technology.

It's because the market is still young. For right now there's money to be made in DHTML controls. As long as that's true, programmers aren't going to be giving stuff away. (Hell, I've got my stash of super-secret components, and I'm willing to bet that you do too.) Once components become more commonplace, OSS libraries will begin appearing.

Re:"Library", are you kidding me?

[Anonymous Coward]

I take it you haven't actually downloaded the code then. There are libraries for animation, DOM manipulation, drag and drop, XMLHttpRequest management and event handling (in addition to the slider, treeview and calendar widgets). That's 30 JS files, not including the examples. That's nearly 10,000 lines of code!

It's fully documented as well.

Re:"Library", are you kidding me?

[Stan Vassilev]

OSS RIA won't be feasible until SVG stabilizes and is as ubiquitous as the Flash plug-in.

I.e. never. SVG has no reference implementations, so in a comparison I saw recently, 10 different SVG "compliant" renderers renders the same content in 10 different ways.

SVG is just a vector engine slapped on top of HTML/CSS and JS. Flash has audio, video, sockets, it has JavaScript 2.0 JIT compiler (coming in a couple of months) with true OOP and it's hella fast.

Flash has advanced authoring environment for building multimedia content and advanced IDE for developers (Flex), SVG has nothing serious. Flash has a strong framework of GUI components (buttons, tabs, windows, checkboxes and all that), SVG has none.

Yea, people do ugly banners in Flash as well, because they can, while in SVG they can't, since there's no support and there are no tools. Also SVG actually can't even do animation (in its pure spec).

Adobe did some extensions to SVG, like add sound and animation, but now that they have Flash, which is better, faster and smaller... well SVG is officially dead.

Oh, and Avalon/XAML is coming. The last nail in the coffin of SVG.

Re:"Library", are you kidding me?

[roman_mir]

There is also Rico. But I think this is all syntactic sugar anyway.

Re:"Library", are you kidding me?

[Anonymous Coward]

However, all the basic functions can be found in Prototype and Scriplicious, which I am sure anyone looking for a Javascript library has already found and is using.

Bloody Breadcrumbs

[nagora]

Oddly enough I was putting breadcrumbs into a client's site when this story came up and I was just thinking "Jesus, breadcrumbs are a total waste of screen space. Why the hell do clients ask for them?" On the other hand, I've yet to find a use for Yahoo! either. Oh, well.

TWW

Re:Bloody Breadcrumbs

[AKAImBatman]

Jesus, breadcrumbs are a total waste of screen space. Why the hell do clients ask for them?

Poorly implemented breadcrumbs are a waste of screen space. Well implemented breadcrumbs are an invaluable navigation tool. They help users track where they are in a given subsection, and help give the user a feel for the overall tree structure of the site. They also improve navigation by allowing the user to reach any level above the current one quickly and easily.

For example, say I'm shopping for a new laptop. I might find myself in the following section:

Store >> Computers >> Laptops >> Bell Demented X

From this point (looking at a laptop) I can navigate back to look at other models, hop up to computers and perhaps look at a desktop instead, or go to the main store and look for deals and other products.

Breadcrumbs are especially useful if I land directly on a page. Say, for example, that someone told me that Company Bell had a good deal on a specific laptop and sent me to the Bell Demented X webpage. If I want to compare with other models, those breadcrumbs help me navigate to other laptops without going through the front page and the entire heirarchy.

Re:Bloody Breadcrumbs

[TheJorge]

While this is well and good, most implementations I've seen are not a site heirarchy as you list, but rather a replacement for the back button. Many times it's a throwback to when the back button would break most apps. One your Bell site, you might be at the Demented X page, but your crumbs show:

Store >> Computers >> Desktops >> Store >> Specials >> Store >> Laptops >> Bell Demented X

And this is actual requested behavior! I'd agree that site navigation tools are important, but the concept of "where have I been recently?" gets taken too far too often.

Re:Bloody Breadcrumbs

[nagora]

If I want to compare with other models, those breadcrumbs help me navigate to other laptops without going through the front page and the entire heirarchy.

I just press Alt-left-arrow and pick a different link; I don't even have to move my hand to the mouse.

Breadcrumbs are an indication that a site is badly designed (or that the PHB/client has seen them somewhere and thinks all "professional" sites have to have them).

TWW

Re:Bloody Breadcrumbs

[AKAImBatman]

* AKAImBatman thwacks nagora upside the head

The back button doesn't work if you land directly on the page. Breadcrumbs also provide information to the user about their location independent from the ability to move to those locations. Pay attention, young padewon.

Re:Bloody Breadcrumbs

[JeTmAn81]

Breadcrumbs are an indication that a site is badly designed (or that the PHB/client has seen them somewhere and thinks all "professional" sites have to have them).

This sounds like the writing of someone who doesn't have a very large userbase for their site, or doesn't much care about providing the best browsing experience for those users. When you're designing a public website, you're aiming to minimize the amount of confusion that may occur on the part of your users. You want to make things easier for them.

Your users may not know to use Alt-left-arrow, and frankly may find breadcrumbs more appealing than navigating through whatever interface their browser uses. Breadcrumbs give you a way of providing navigation for your users independent of browser type.

Re:Bloody Breadcrumbs

[nagora]

This sounds like the writing of someone who doesn't have a very large userbase for their site, or doesn't much care about providing the best browsing experience for those users.

Help! Help! I'm being attacked by a usability-fad-of-the-month nazi!

TWW

Re:Bloody Breadcrumbs

[Anonymous Coward]

I recommen you pickup a book about HCI before shouting things you really don't have a fucking clue about.

Re:Bloody Breadcrumbs

[eMartin]

The closest thing I've EVER seen to that is progrees indicators for web shopping carts, and that is mainly there to show you how far you've gotten in the process.

In fact, I'd go as far as to say that what you've described isn't breadcrumbs at all, which were really another way of displaying a "file path" style view of where you are on heirarchical sites, such as web forums or stores with categories like the GP suggested. That they are clickable is a bonus feature taken from file managers that added that.

Re:Bloody Breadcrumbs

[lamz]

Do you have an example for that claim? I've never seen anyone do that.

Re:Bloody Breadcrumbs

[nagora]

I recommen you pickup a book about HCI before shouting things you really don't have a fucking clue about.

Users don't learn what sites to like from a book. As a user I find breadcrumbs pointless. Sorry to have disturbed your dreamworld; go back to your books.

TWW

Re:Bloody Breadcrumbs

[kchrist]

That's an interesting reaction to someone who's only advocating making things easier for the end user. When designing UIs, what you find easy is only important if you're the only person using the app. If you expect others to use your system, don't you think it wise to find out what they prefer and expect?

Re:Bloody Breadcrumbs

[nagora]

When designing UIs, what you find easy is only important if you're the only person using the app.

I'm not talking about when I'm designing; it was just a coincidence that I was typing up a site with breadcrumbs when the story appeared.

As a USER I find the current trend of having breadcrumbs inexplicable except insofar as it seems to go hand in hand with sites which are thrown together and then have some navigation trick stapled on in order to avoid the whole plan being totally unusable. In other words: as a USER my impression is that breadcrumbs are generally used by lazy designers to avoid actually thinking about their sites' designs and logic.

Sheesh.

TWW

Re:Bloody Breadcrumbs

[kchrist]

As a designer, what do your usability tests tell you?

Re:Bloody Breadcrumbs

[nagora]

As a designer, what do your usability tests tell you?

That people react positively to the sight of a trail, and say they like them but when their usage is monitored they actually rarely use them.

It's like clients who ask for - nay, insist on - a CMS for their site. They love them. They pay extra for them. They never fucking use them! We even tell them: "look, you want a content management system; everyone does. But your site is pretty static; it would be cheaper to just pay us once a year to update the page that holds the staff biographies, because that's all you're ever going to change." But, oh no, they have to have it. Then a year later they phone up and ask if we could make some changes to the staff page. People, huh?

Which is why I end up making sites with trails and grumbling about it.

Re:Bloody Breadcrumbs

[JeTmAn81]

Users don't learn what sites to like from a book. As a user I find breadcrumbs pointless. Sorry to have disturbed your dreamworld; go back to your books.

I don't think you understood kchrist's point. The idea is not to design what you as a either a designer or a user would find most useful and efficient. The point is to design what your users (again, not yourself) will find most useful and efficient. In this case, you say your users don't use the breadcrumb trails, and that's fine.

But on the site I work on, our users DO use them, and find them very handy. I hear about it rather quickly if the breadcrumb links are dead or aren't up to date. So it's a good thing to have them there, as it makes the experience of browsing our site easier and more suitable to our users. That's why I object to your blanket condemnation of the use of breadcrumbs and assertion that they are an indicator of "poor design".

Re:Yahoo, giving guidance on Web design?

[CynicalGuy]

you can completely customize every aspect of your Yahoo portal, and put drop RSS blocks onto it.. if yours is cluttered and useless, it's because you're too dumb to change it..

Re:Yahoo, giving guidance on Web design?

[Anonymous Coward]

Yahoo has one of the least useful, most cluttered, and unappealing sites on the web.

Well, they -are- big, and their -whole- business -is- dealing with the web... so maybe they do know something about making money using this Internet thing... thus, when they offer some tips, why not just listen and learn as opposed to criticize a relatively successful corp? (their site may be cluttered, but they're pretty damn good at what they do).

now the web 2.0 can take off!!!

[mookie da wookie]

Now the wonderful Web 2.0 can really take off! I am so looking forward to a Service Oriented Architechture where Web Services can get a ground level understanding of my Ad Hoc Supply Chains! Why, I may design a Workflow to help me document and process my well-formed giddyness! Fabulous!

Re:now the web 2.0 can take off!!!

[BluesMoon]

you ought to write this up as a template for the nonsense generator (search freshmeat).

Re:Yahoo, giving guidance on Web design?

[g051051]

What do you mean, "your Yahoo portal"? Why would I want to or need to customize their site? You seem to be agreeing that the default page presented by Yahoo is cluttered and useless. Considering the useless mess I'm presented with as their front page, I have no interest in proceeding any further into Yahoo land.

Oblig. Grinch

[neveragain4181]

I for one welcome the new script (overlords), but I can't help myself (oh, and I tried) but point out that because of the nature of the technology (client side jscript) then there isn't actually a good way *not* to release the source code for re-use, at least for the Beer part.

Now I do think Yahoo has done a smart thing in doing this under a BSD license, but it's worth remembering that this might be because they don't really have a way to protect their IP anyway. You can muss up script to be less readable, but basically it has to execute and therefore scarf-up-able to those that want it.

If this was a server-side technology then I don't know if Yahoo would have been so willing to go both kinds of free? At the very least this messes with Microsoft's Atlas people's heads, so should be good to sit in the peanut gallery for this one.

Am I being too cynical for a Tuesday?

N/A

Re:Oblig. Grinch

[AKAImBatman]

I for one welcome the new script (overlords), but I can't help myself (oh, and I tried) but point out that because of the nature of the technology (client side jscript) then there isn't actually a good way *not* to release the source code for re-use, at least for the Beer part.

Yes there is: It's called copyrights and patents. You can easily have your ass handed to you on a courtroom plate if you try to rip off someone else's JavaScript. Make sure you have permission before you start stealing code from others. If you're unsure, do your own implementation. You may still be liable for patents, though.

Re:Oblig. Grinch

[Anonymous Coward]

They have copyright law and any patents they may hold to the technology on their side to enforce their rights. About the only thing they can't claim is trade secret since the code is client-side.

Re:Oblig. Grinch

[FooAtWFU]

That's not really a good way to not-release-the-source. It's a rather bad one, really. You need to track down the offenders and take them to court (hope they're not overseas!) and there's always the risk of some random using it somewhere or modifying it in a way so you can't detect it...

A good way to not-release-the-source is some sort of binary distribution that cannot be trivially decompiled. To some extent, you can obfuscate your JavaScript pretty well with various tools that are out there, but if it's meant to be a usable API or toolkit then you can't go renaming all the functions to an unreadable mess- or what's the point of the toolkit, anyway?

Re:Oblig. Grinch

[AKAImBatman]

A good way to not-release-the-source is some sort of binary distribution that cannot be trivially decompiled.

No such thing exists anymore. Even complex C/C++ programs can be decomplied quite easily. It's just that people feel more exposed with JavaScript because the code is distributed in source form. Try a packer/obfuscator like this one instead. The code can still be extracted by someone smart enough, but it does put a barrier in their way.

Re:Oblig. Grinch

[neveragain4181]

Yeah, ok - fair point (am I allowed to do that here?).

Have to say though that patents for a calendar and tree view javascript (+2 years from now) seem a little unlikely. Not saying someone hasn't carpetbagged them already, but Yahoo take the PR to fight that case? Seems unlikely.

Also, the copyright may be more easily enforced, but really would it take too much to tweak them a little - that's basically what people do anyway with script (so I heard from a friend of a friend, can't remember his name...).

Also, anyone know of a copyright case that went against the defendant where a treeview or calendar javascript got stolen? Or maybe cite a similar case, i.e. a cookie crumb trail? It seems similar to people copyrighting their html and css, but then others doing a View Source and editing for their own use/learning and producing derivitives - do companies sue for copyright on that? This hardly a lot of difficult to do software...

Anyway, I do agree with your point, but it seems unlikely, which is why I think this is mainly for PR due to the nature of the tech. My point was that if they could enforce their IP they might not have gone this way.

N/A

Re:Oblig. Grinch

[AKAImBatman]

fair point (am I allowed to do that here?).

I always do. Trolls don't like it (it smells too much like a real conversation), but it works well for those of us who like to actually discuss things. ;-)

Anyway, I do agree with your point, but it seems unlikely, which is why I think this is mainly for PR due to the nature of the tech. My point was that if they could enforce their IP they might not have gone this way.

It really depends on how valuable the technology is to you. If I implement a gaming library in AJAX, I'm going to be pretty peeved if someone rips me off. Peeved enough that I will probably sue for lost revenue. On the other hand, if someone rips off my cool looking button control, I'll probably just let it go. (Not worth it.)

Now if a competitor develops their own clean-room gaming library, then we have market competition and there's not a whole lot I can say about it. Unless, that is, they happen to step on a few patents I've filed for various techniques. Then I can probably extract a revenue from them even though they're a competitor.

Please don't use the drag and drop

[slashkitty]

It breaks things on webpages and is really pissing me of on their my.yahoo.com site. If you don't need to drag and drop things, why have them? If you don't need to open a page in a new window, why do it? I'm starting to really hate some of this AJAX stuff.

Re:Please don't use the drag and drop

[Anonymous Coward]

ajax is using 7 year old techology circa windows 98 what do u expect...windows 98 performance?

Re:Please don't use the drag and drop

[DarenN]

I hate the blink tag too, but don't knock HTML, I just want to go on a rampage and kill the developer and everyone involved in the site. The moral of the story is, blame the developers, not the framework/language. AJAX has some great ideas. It'll take time to mature, though!

Re:Please don't use the drag and drop

[Cocteaustin]

If you'd actually RTF-Design-Pattern-Library, you'd see that the recommendation is to use drag/drop in addition to accessible methods of interaction, not as the sole method.

this is not a widget library

[Gopal.V]

The important peices are *NOT* about widgets. This is about the ygPos,ygAnim and ygDom libraries which are invaluable to most people (at least me).
The animation systems are actually pretty awesome. The cacheTween() functionality in there takes it very close to what I've been doing with flash previously.

Morover, Y! has been using these for the past 6 months on different browsers before they open sourced. That part is really what most people look at.

Re:this is not a widget library

[AKAImBatman]

Same advice applies.

This is about the ygPos,ygAnim and ygDom libraries which are invaluable to most people (at least me).

This is what I love about doing DHTML/AJAX. People are so easily impressed. :)

Re:this is not a widget library

[wumpus188]

Have you seen prototype.js 's sister Visual Effects library?
Yahoo's stuff looks pretty weak in comparison... not to mention that VisualEffects uses much more clean implementation with prototype.js

Re:this is not a widget library

[Anonymous Coward]

Prototype adds additional members to the Object prototype in JavaScript, which breaks core language functionality (for var i in obj) and doesn't play well with other scripts. The Yahoo! stuff doesn't do this.

Yahoo!'s animation library is in a different class entirely from script.aculo.us . Script.aculo.us has some neat built in effects, but the Yahoo! thing can do multi-control point bezier curve animations with the path pre-calculated for performance. It's a very powerful tool.

Re:this is not a widget library

[slashddot]

All of the functionality offered by ygPos, ygAnim, ygDom, etc. has been available for a long time elsewhere.

Prototype is a de-facto standard set of extensions for JavaScript and DHTML. MochiKit is another powerful library. Projects like Dojo, Rico and Script.aculo.us build on top of those libraries to privide UI functionality like non-HTML widgets, grag/drop and animations.

There is very little real value in the code Yahoo! just offered, unless you want another library to do the same things. This is nothing more than a publicity stint.

Comments interesting and appreciated...

[ursabear]

I sincerely appreciate the comments of those who know (much more than me) about web UI, techniques, technologies, and patterns. As a server-side engineer and developer, I don't spend a lot of time on the front end. It's nice to see how /.ers digest this type of information and re-present it from lots of angles.

With that said, I'd also like to say that the pages are pretty well done. It is obvious that a great deal of time and effort was spent conceiving, writing, and, producing these beginnings of libraries and instructions. I found the effort to be commendable and interesting.

For someone like me, these types of efforts actually help me understand quicker and keep me interested.

Yahoo UI Blog

[Anonymous Coward]

http://yuiblog.com/
The blog to go with the (future) releases.

Re:Yahoo, giving guidance on Web design?

[CynicalGuy]

http://my.yahoo.com/ .. make it look however you want.

UI Controls

[int19h]

The UI Controls all looked nice and worked very well in Firefox 1.5.
Unless there's a Grand Hatch (tm) that we've all overlooked, I'll start using this right away*!

So, thanks Yahoo.

*) In this particular comment, "right away" is defined as "tomorrow, or any other time I feel like checking it out".

Re:Yahoo, giving guidance on Web design?

[g051051]

Again, what does this have to do with the atrocious look and feel of www.yahoo.com?

Earth

[Anonymous Coward]

Call me pedantic, but it's "Google Earth", not "planet earth".

Re:Yahoo, giving guidance on Web design?

[generic-man]

You're judging a massive company by the way its end-user home page looks. Other people are trying to explain that Yahoo! offers other interfaces and techniques for accessing search. You're ignoring them and continuing to whine about yahoo.com, which you don't have to use to get to their search engine. Did someone set your homepage to yahoo.com as a prank or something?

Re:Yahoo, giving guidance on Web design?

[Brad0415]

How the hell do you expect *any* company to appeal to everyone's individual sense of style and aesthetics, not to mention content? Give me a break. If you are too lazy to customize the site then that is your problem, not Yahoo's. With a very minimal time investment, you get what I believe to be the best portal/mail combination out there (using their beta mail).

Re:blah...

[Anonymous Coward]

You must be new here!

Re:Yahoo, giving guidance on Web design?

[g051051]

I'm judging the applicability of Yahoo offering any sort of web design guidance when they can't even produce a decent landing page for their site. As a developer, their cluttered, ugly, messy web site is a serious negative influence on me taking anything they say about web design seriously.

And who's even mentioned search? You simply created made that inference up out of thin air. Read more carefully before you post.

Re:Yahoo, giving guidance on Web design?

[g051051]

Again, why should I need to customize their web site? Why do I need to invest my time in cleaning up their poor design, ugly layout, and otherwise atrocious web pages? Yahoo has traditionally hasa cluttered, unappealing design, and it continues to grow worse as time goes on. And if that is their idea of good web design, I'll have no part of it.

Re:Yahoo, giving guidance on Web design?

[Forbman]

So, they give you a chance to make it how YOU want it (my.yahoo.com), and yet you still bitch & moan about how bad the default yahoo.com is.

Shut up already.

Re:Yahoo, giving guidance on Web design?

[generic-man]

Yahoo! is offering tools to add all sorts of funky JavaScript shit to your web site. How you design your site is up to you. You can have a minimalist interface or a maximalist interface, or something in between. You seem to be making the conclusion that Yahoo! is advising you on the proper layout, CSS, and advertising format for your web site; that is not the case.

Thank you for marking me as a "foe," though. I look forward to not reading your reply.

I wish I had these 3 years ago

[josepha48]

.. Then I would not have created my own client side calendar.

I think my calendar is easier to use. And mine creates the div and checks to see if it already exists.

Someone now needs a UI Design tool that allows people to easyly integrate these into a design WYSIWYG mode.

Calendar foo = new Calendar('your_id');

Not all their APIs are the best

[Wee]

Very true. Their developer APIs are the best of any major offering.

Not really. Have you ever looked at their ads API? It can't even bgein to compare to what Google offers.

It takes about 90 seconds to sign up and start getting access to advertising data via Google's API. It's SOAP, so pretty much every programming language besides BASIC and Forth are supported. Google has loads of documentation online regarding their ad API program. And it's free. You get to do what you want with the data that you get back.

Yahoo has had an advertising API for 5 years now, but what does it do? What does it take to get access to it? We know it uses REST, but what data can you get back from it? How much does it cost? Where is the sample code? Is there a support forum where I can talk to other developers? What are the terms of service? Can I use it to get 3rd party access to others' data? Are there any other restrictions on using it?

The API page linked above doesn't answer any of those questions. Hell, the ads API isn't even listed on the developer resources page you linked to. Why is that?

So if I email xml-ysm@yahoo-inc.com and ask the above questions, how long before I get a response? Will all my questions be answered, or will I get more questions back then answers? Try it. I did last spring. It's an interesting response to say the least.

You're right in that Yahoo has some very nice developer resources. But this is one area where Google substantially outshines Yahoo. Send a short email to the above address and ask to get access to the Yahoo ad API. Seriously. Just send a one-liner. Take a moment to read through the canned response you get back. And then ask yourself "Why don't they just put all that info on a web page somewhere out in the open?" That you even have to email someone to begin with is odd (and annoying). What is Yahoo hiding? Why are they being so cagey?

Compare that email response to the AdWords API page at Google. Now step back and take in the Yahoo ads API page (and, I suppose the one other page regarding their API). Add in the email repsonse. Now take in all the info on Google's API. Notice a difference? Just a small one, maybe?

There's just no comparison whatsoever. Google is open, free and easy with their ads/cost data sharing. Yahoo is, to be kind, not so much any of those things.

Anyway, even if you DID manage to get API access, you better not hang your hat on that since access could get pulled for any number of spurious reasons. Take a look here for an interesting read. Yeesh.

Yahoo itself might have a decent suite of APIs, but such notions haven't as yet affected the folks who came over from Overture.

-B

Re:Not all their APIs are the best

[stupidfoo]

Ok. So it doesn't have an ad api that's up to snuff. That still doesn't negate my point that, overall, they have the best offering.

Re:Not all their APIs are the best

[Wee]

Well, I was taking each API they offer on its own. The Yahoo mapping/travel planning offering might be the best of the bunch, but the ads one definitely isn't. I suppose if you grouped them all together Yahoo would come out on top.

I guess it depends on what's important from a developer/business point of view. A lot of money can be managed via an ads API. Having one vs. not having one can make a huge difference in ad spending for some organizations (SEMs, especially). But if you're a web developer looking to do something CMS-ish or whatever, then the AJAX stuff they just released is more important to you.

-B

Just FYI...

[Anonymous Coward]

This library comprises a number of dynamic HTML utilities...

This library is composed of a number of dynamic HTML utilities...

-SNS

Opera Users...

[Egataes]

...can't access a number of Yahoo! pages (nearly anything apart from their search page) due to a browswer restriction on Yahoo's end. I wonder if this extends to their other material.

Re:Opera Users...

[XL70E3]

I'm using opera for as long as i can remember, and never had any problems on yahoo pages, and believe me, i use yahoo pages extensively. I just don't see what you are talking about here!

Design Tools? Bah!

[kerouacsgp]

Design Tools? Bah. Real men code by hand!

Overrated

[Anonymous Coward]

Doesn't even bother to link to a single one of the so obviously superior alternatives.

Good compatibility too, this is great!

[lux55]

Looks like at least the slider examples all work in Firefox, Safari, and Opera for me. This is great news!

I think a standard JS widget library is a definite good thing, and with a company like Yahoo behind it it will probably get more developer awareness and traction with the business folks too than what's been available up to this point from smaller developers.

BSD licensed too, so it's free in any sense of the word for any use (good for us half-commercial folks ;)).

Yahoo found a nice way to do js namespaces

[Nathan Cassano]

Yahoo has found a nice way to do namespaces in Javascript. I think this could be used more widely by other developers.

Code:

YAHOO.widget.TreeView = function(id) {
    if (id) { this.init(id); }
};
 
YAHOO.widget.TreeView.prototype = {
    id: null,
    init: function(id) {
        this.id = id;
    },
}

Example:

objTree = YAHOO.widget.TreeView(1);

So.... this is newsworthy?

[Anonymous Coward]

Sorry, but I've been using the Backbase library for a while and Yahoo's stuff just isn't even close. I feel like I'm reading a thread about the benefits on attaching stone wheels to a wooden cart.

As I was glancing through the Yahoo docs I kept wondering... where is the XPath support for selecting objects, can I create HTML templates for my own custom controls, does it include lots of sample code/demos/pdf tech docs?

Also, what about a full-featured client-side debugger so I can see a history of AJAX calls (request/response), the current DOM tree in memory, and test out XPath queries and see the results?

Not even close... not even close...

Re:So.... this is newsworthy?

[nate nice]

Backbase also ins't free. There is a "free" version but if I do recall you cannot use their free version for any commercial Website. But then, their API is feature rich enough if you need this ability it may be worth the money to invest in their API.

Re:So.... this is newsworthy?

[SSpade]

Is that the same Backbase that has "Your web browser is not compatible with the primary Backbase site. Find out which browsers are supported at the Compatible browsers page." at the top of the page, and whose demos are entirely broken?

Backbase support for Safari and Opera has been "forthcoming" for quite a while. It's not there yet. Which means that Backbase is about as useful as ActiveX for public websites.

Nice architecture. Once it actually works I'll pay more attention to it. Right now? Not even close.

Damn...

[trevdak]

Damn... ust after I had reverse-engineered and commented Google Suggest

You think I kid.

Re:Yahoo, giving guidance on Web design?

[g051051]

What part of the JavaScript offering is even under discussion here? This is entirely based on the Yahoo! Design Pattern Library. None of this pertains to the hypothetical possible theoretical cahnges that you may or may not do at a different site (my.yahoo.com), but the atrocious www.yahoo.com. Why can't you get your facts straight, and address the question at hand, rather than constructing imaginary arguments, attributing them to me, and then rebutting them (poorly, I might add)? You're just talking to yourself.

Horay for Yahoo!

[Builder]

Isn't this great ? Yahoo! are giving us Free! Code!

Pity about the reporters they've shopped to the Chinese... They're probably quite uncomfortable right now. But hey, Free! Code!

Re:Yahoo, giving guidance on Web design?

[g051051]

Who cares about my.yahoo.com? I'm speaking of the default, first thing a user sees, first impression of yahoo: www.yahoo.com. This is supposed to be their best face, the thing that attracts people to and keeps them at their site. Instead, it's a horrible mish-mash of poorly organized, random content, the visual equivalent of cacophony. Why is it MY job to make their site useful?

Very Promising, A Little Amateurish

[psydeshow]

Great libraries, beautiful examples. But perhaps they are still working on the Yahoo! Code Forge where these "libraries" get version numbering, changelogs, full API documentation, bug reports, etc.

Maybe nobody else cares about these things the Javascript, but they would never release server-side code in a generic, rootless ZIP archive. So much promise here, but I'd wait to see some better change management before integrating these libraries into a production site.

(Full speed ahead on development, though!)

Re:Very Promising, A Little Amateurish

[BluesMoon]

The API docs are in there at least.

Stocks, Geocodes, Local XML

[lababidi]

Yahoo offers a lot more than google in certain aspects. Yahoo provides Geocoding ability to turn addresses into latitude and longitude coordinates in an xml file format. same with their local search.

they also produce a CSV sheet for stock quotes.

but google does have the advantage in interface. google maps does beat all competition for example.

Too Bad Yahoo is Evil

[DirtyLiar]

Sounds like a nice set of tools.

Too bad Yahoo is Evil.

http://australianit.news.com.au/articles/0,7204,18 109221%5E15841%5E%5Enbv%5E,00.html

http://www.google.com/search?hl=en&hs=9r4&client=f irefox-a&rls=org.mozilla:en-US:official&q=yahoo+ch ina+dissident+email&spell=1

Re:Too Bad Yahoo is Evil

[DirtyLiar]

Sorry, sould have been more specific.

Yahoo helped the Chineese government track down and jail a dissident.

Re:Yahoo, giving guidance on Web design?

[pete6677]

This page, when filtered through an ad-blocker, looks quite nice and has a lot of useful information. It always throws me for a loop when I go to Yahoo on a friend's PC with no adblock; I'm amazed at how much clutter there is.

try again

[Anonymous Coward]

You're completely missing the point of what a design pattern is. A "Design Pattern" as you so charmingly capitalize it is NOT specifically this or that problem domain (it was ripped off^W^Wappropriated by the GoF from an architecture book that was over twenty years old, for christ's sake), but rather a set of common problems and solutions in any given problem domain.

So "UI idioms" are just as deserving of an attempt at the development of a pattern language as, say, object oriented wankery or building construction. Elitism about "my problem domain is more academic and abstract and uses more ten dollar words than yours" is lamer than lame.

Very Impressed!

[Morden]

The Design Pattern library is a fantastic resource for developers who want to do the 'right' (well, by Yahoo standards, anyway -- I find it very sensible at least) thing with their sites' user interfaces.

I've forwarded it to the designers here at work, hopefully I'll start to see some good come of it in our interface design :)

It's seriously great to see this kind of thing happening.

Hate to tell y'all this.....

[noirsoldats]

.... But Google's had their AJAX Libraries opensource and available to the public for a LONG time (http://sourceforge.net/projects/goog-ajaxslt/). So Yahoo just got religated back to second-class. Nothing more to see here, move along. I for one, welcome my new google overlords.

Re:Yahoo, giving guidance on Web design?

[Anonymous Coward]

http://search.yahoo.com/ :-)

Yahoo Avatars

[Anonymous Coward]

This guy has recently found out a easy way of getting Yahoo avatars from yahoo servers directly into your webpage/blog! He has put up some samples here. He has also put up a complete article on how to do it here.