Yet Another Blog from Luar

If someone ask me, he wants to build a public-facing application, for SEO problem, how to judge whether he should use Flex to build it? I will not tell him Flash has method to solve SEO problem in dynamic, data-driven application, it is a lie because it is possible but hard to do it. I will use a very simple principle to jugde. Although your application is public-facing, but is it the features of application can be used only after user login? If yes, no need to worry about SEO problem, it is a business logic / security problem over SEO problem! Go ahead with Flex.

Browse MXNA, know Adobe exchange have convert to HTML version, another good act after Edge newsletter convert back to HTML version. It is not only me happy on this new (1, 2). This is not a bad symbol for failure of SWF application but a good symbol for give up all bad usability applications! I always noted in the past, build a HTML-like SWF application, cannot gain any great user experience but lost all benefit in a simple HTML text application, it is not worth to do, it is not a wise decision and a hurt to user. Users hate bad usability HTML application, users hate bad usability AJAX application, but users hate bad usability SWF application most, because is it possible to use Greasemonkey to improve SWF application?

Leave comment for the thread in Adobe forum about the new HTML version

Like many Adobe developers, I check MXNA daily, in order not to miss any useful information, I scan through all titles and click the topic I am interested, but as the aggregated blogs increasing, off-topic articles increasing and every week must have someone flooding MXNA with all his old post, I am wasting time to filter out these unrelated topics manually, I have to use smart category finally. I am interested in all Flash related topics: Flash, Flex, Flash Lite and Apollo. But smart category is broken Flash topics to several different areas: ActionScript, Apollo, Flash, Flash Mobile, Flash Platform, Flex. I have to check them all, sometime I find the topic is cross appeared in different areas, but some topic is not. Therefore, my time is still wasting in filtering duplicated topics.

What I suggest: MXNA should release a more clear criteria what keyword is looked for in different smart category. I also expect smart category “Flash Platform” should covered all topics in ActionScript, Apollo, Flash, Flash Mobile and Flex.

• 给各位疯狂迷恋技术的人提一点建议..
• 您的应用是否真的需要Flex？ by Flex Builder Linux version hacker
• 小菜编程成长记（六 关于Flex的争论）

Main concern: what is a page-based application? what is a screen-based application? Is it every task can be done in the same screen? Using MDI (Multiple Document Interface, aka windows)? For a large enterprise application with several features, how to design it well in a screen instead of separate it to several pages?

When SWF application cannot natively, easily support for back button, deep linking. Flex zealot will argue it is an application, it is not a website, no need for care about such issue. Why users still ask for such requirement? Because even Flex developers are still stay on the concept of “page based application” instead of “screen based application”.

Suppose an e-commerce site administration system, it has three features, seller can add products, review the orders, update shop information. A typical web based application will consist of a menu bar with three buttons to click, then the whole screen will use for one feature operation. For the user, it is no differences between it is a HTML/AJAX/SWF application, one screen does one task, if he clicks a wrong link to another feature screen, he will immediately click back button to existing screen.

If his daily work is using “review order” most frequently, he wants to bookmark “review order” screen. When he enter the page, application know he is not logined, ask for login first, then redirect to “review order” screen.

It is already a common practice on how we develop old HTML page based application since 10 years ago. Come to 2007, what we saw the Flex UI part, it is still act as a HTML Form element replacement. If Adobe / Flex Evangelist cannot work out a best screen-based Flex application user experience design patterns / examples / show cases (not that little flickr app or dummy mobile phone shopping app), Flex developers will still using their old school “page based” concept to develop the application. Back button, deep linking curse will be a never ending nightmare to Flex and Adobe.

Other Reference

• Saving User preferences on the server – cutting corners in Flex

In my Chinese version blog, there is a voting conducted for a period of time (since last year September), so far have 700+ voted from (China, Hong Kong, Taiwan).

What is your current RIA Technology used?

1. Ajax
2. Flash
3. Flex
4. Ajax+Flash/Flex
5. Still using page based refreshing HTML

Counted to today, The result is interesting (note: it is not a scientific research or any official statistic, it is just my personal blogger readers voting): Flash is the majority but Flex is the minority.

My own opinion is, since my blog is deicated to Flash, most of the readers are already Flash developers, so the usage of Flash should be high, people are already aware used SWF as RIA technology since Flash MX and started as a main stream solution, but why the adoption rate is so low in Flex? I think it is due to lack of learning resources: Flex resources are written in English and it is a language barrier for developers adopt it. So what should Adobe do to explore a big potential market in this region, I think they know how to do.

Main concern: No offense here, I just find the comments are funny and interesting, just want to quote in my blog.

James Ward’s article “How I Overcame My Fear of Flash” is another marketing article for people who do not understand Flash, worth checking on its comment below instead of the article body. For a SWF-RIA technology promoter, how can you answer the following quotes:

Adobe took a cross-platform IDE (Eclipse) and turned it into a Windows-only application (Luar’s remark: it has MAC version)

And on the subject of search-engine indexing, this is a huge issue that you largely ignore. Or better said, you pass the responsibility to the search engines. If Flash were a truly open format, then I would somewhat agree, but it’s not. It is proprietary (whether you excuse it’s proprietariness or not). Take, for instance, the company I work for. They have dictated a non-standard font for all main page titles. As a result, the title must be done as an image, thus ruining indexability and driving down page-rank. Would you suggest Google must check all images, and learn to index those few images that happen to hold a raster-version of a word?

I’m personally hopeful that Sun is successful with open-sourcing Java, and that drives Adobe to follow with Flash (the format and player)

For me, any flash site that attempts to do the whole experience in one flash movie, runs like a dog on my laptop. Flash is nice for games, artist’s websites etc.
But NOT for web applications. Just too heavy, takes too much ram, and is plain annoying. I use the Flash 9 player for linux, and although YouTube works fine, not all sites do and sometimes i get audio in slow motion. Html is standard. You get what you expect to get, forms behave the same etc. Modern Javascript are no longer a pain to write and you can do a lot of cool cross browser things with JS.
Flash has its place, but not as the be all end all web UI. it’s just out of scope for that i think.
Harel

I don’t plan on writing internet Flex apps for the same reason that I didn’t write IE-only sites from 2001 to 2005. I’m against any entity — commercial or no

Flash support for Linux is not the issue, it’s just a symptom. The fact that Flash support for Linux has improved somewhat recently in no way addresses the real issue and the root cause of the problem: as a closed, proprietary standard Flash is *hostile* to the entire basic principle of the Web. The whole concept of “support for Linux” (or any other specific OS) is an insult to the Internet in itself.

The attitude of Adobe/Macromedia towards open standards is more arrogant and destructive then Microsofts’ ever was. Microsoft is quickly coming around, but Adobe still believes it can hold on to its RIA monopoly at the expense of everyone else.

Although like most other developers I’m forced to use Flash for lack of a better alternative, but I can assure that if such an alternative appears we will drop Flash in a heartbeat. Even if Adobe opens up, it will be too late. Right now many of use prefer to use AJAX where we would before use Flash, not because its easier (god no) but because it gives us the option of creating the right user experience without selling out to Adobe.

I’m as pragmatic as the next professional when it comes to using proprietary stuff, but please don’t defend Adobe’s policy as if it were something to be applauded. Adobe’s attitude towards the internet and the necessity of open standards is as negative and arrogant as Microsofts’ was in the nineties. If Googles’ guiding principle is “don’t be evil” and Microsoft currently seems to go for “try to be less evil”, Adobe/Macromedia still firmly sticks to “be as evil as you can get away with”.

P.S. I think as a Flex Evangelist work for Adobe, better post more practical Flex development articles/case studies (stronger existing Flex developers skillset / Flex-interested user confidence) instead of marketing towards outside people.

Main concern: most outsiders challenge Flex: I can create the similar site with AJAX/HTML/CSS, why I have to use Flex? I do not want Flex to create HTML-like application, but turn out too much such Flex applications in the market already, why Flex developer cannot find a new mindset to design a good RIA? How Flex solve the UI problem differently that HTML cannot solve?

My doubt series have not finished yet, but I am too busy on my working project. I saw the blogsphere had started concern on what I thought, and they have good analysis on this topic already:

• One of the nicests Flex apps so far
• Overcoming design laziness
• Do all Rich Internet Applications technologies suck?
• Is RIA design degenerating into ‘skinning-the-functionality’?
• Is RIA design degenerating into ‘skinning-the-functionality’? (Take 2).
• But It Looks Too Much Like Flex
• Site Smackdown: Flash c. 2000 vs. AJAX c. 2006
• Is Technology Really the Answer to the UI Problem?

When Macromedia promote RIA since Flash MX, their own site become one of the experimental place for their own dog food, e.g. the only full Flash version corporate site (appear for a very short period, and convert back to HTML later) and exchange. What I saw is a HTML style RIA, we, and our client, and our client users will challenge, why this “html” site breaks my browser function? If we use Flash to reproduce a HTML site without any gain in experience or functionality, but losing lot of usability. It has no strong reason to use Flash.

For what I saw and above bloggers saw, Flex is just a tool for replace HTML style applications. We worry to see lesser innovative use in Flex application after Flash Golden Age, check FWA: The 10th Anniversary of Flash, too.

But, we see Ajax dare to try new, dare to innovate now (OK, you can said they are copy old Flash style), I see innovation on Ajax is coming more but on Flex is coming lesser, except fewer pioneers.

Main concern: I want native print preview support in Flex.

When we talk about usability problem in Flash, you may expect I will repeat the back button support, deep link, save web page problem, SEO (1, 2, 3, 4) in Flash, of course not necessary. Ha! I understand we should talk in the same domain, use Flex to build “enterprise application”, “Intranet application”, which are not public facing, consumer website. We do not need to care about back button, SEO… those problems.

Starting from Doubt on Flex as the best option (3), I will focus on the usability problems. Yes, Flash usability problem can be solved but hard to solve, when you gain productivity in Flex framework, Flex builder, if you put in time in solving usability problem, you will loss more time than you gain.

Let start…

One of important feature in enterprise application is printing: print report, print customer record.

Lack of print preview
In Flex, we can use print-specific child component to design what should print on a paper, which may different from what is shown in the screen. However, once the user click print button, a printer dialog appear, what he can do it print it immediately, he cannot preview what will print on paper.

Lucky to MAC user, MAC OS print provides a preview feature no matter what application you are using.

Lucky to HTML/AJAX based application user, the browser help you, the browser help the developer,too. It has a print preview natively support for any web page.

Flex has provided many visual UI element you can use them easily but lack of a strong integration with OS feature, what will happen when the application move to Apollo? Can user treat Apollo application like traditional desktop software, enjoy the existing, standard feature they have in other software?

In short term, we have to create our own print preview feature in the Flex application, in long term, Adobe must provide a print preview function natively work with OS, not just a simple PrintPreview component.

Is it possible to liquidize layout of print content in different oriented paper in Flex?
Another feature we are existing enjoyed in HTML/AJAX based application, inside print preview, user can choose landscape or portrait, the print content will liquidize to fit the paper width, it is not what Flash does scale to fit the paper width. It is a two lines wrapped tabular data in portrait and become one line tabular data in landscape.

Follow up

• Doubt on Flex as the best option (4): Dilemma on Follow HTML style

Other Reference

• FLEX: How to Print an Invoice

Main concern: usability is the most important part in RIA instead of (fancy/rich/refreshless) user experience.

Jakob Nielsen’s article Flash: 99% Bad liked a big rock throw into the dead water, created a shock among Flash community at 2000. Later, Macromedia invited him to develop best practice guidelines for creating usable RIA with Flash. What he returned to our Flash community is one Developer Center article only: Usability Review of the Pet Market Demo. The true answer for Flash usability is after, a PDF for Flash Usability: Design Guidelines for Web-Based Functionality, Tools, and Applications, you have to pay for US$64 to buy it. Yes, I have bought it at that time, because I am really care about usability and want to learn a lot on it. We seldom hear about his comment on Flash usability again.

I understand criticize on Flex does not mean AJAX is good (vice versa, but s(a/o)me Flex zealot cannot understand), when we saw AJAX has 100 problems and Flash has 50 problems on developer side and user side. What is the point for you make a conclusion Flex must be better? Starting from Doubt on Flex as the best option (3), I do not focus on comparison between them, let talk about Flex itself, the usability problems.

Flash usability problem can be solved, but it is a problem developers not aware and not willing/have time to solve. Yes, Flash usability problem can be solved but hard to solve, when you gain productivity in Flex framework, Flex builder, if you put in time in solving usability problem, you will loss more time than you gain, therefore, is it Flex a worth invest development tool and SWF is a right medium for RIA?

When you choose “My most used RIAs”, can you even choose most used Flash RIA? Sorry, I cannot choose (except YouTube and Breeze <- but it is not daily use), I do not like to use SWF applications, 99% are bad in usability. SWF developers who do not eat their own dog food cannot understand users are suffering when using bad usability SWF applications.

Follow up

• Doubt on Flex as the best option (3): Bad printing experience

Others’ Article

• Seven ideas for building better RIAs
• Creating Passionate Users (A blog dedicated to user experience, must check)