The company’s reasoning for endless beta cycles has never been clearly explained and Google always insisted the tag would be removed once the product was ready. I suspect Google’s motivation for beta tagging is a combination of:

• it reminded users that the web application was experimental and undergoing development
• it lowered user expectations of early web systems and gave us a pleasant surprise to find fully-functional online products
• version numbers rarely have much meaning when applied to web applications and services. Unlike shrink-wrapped disk-distributed software, web solutions can be incrementally improved on a daily basis. Few people, other than the developers, will care if it’s version 7 or 57.

However, Google has finally recognised that the “beta” label has different connotations in the business world and it could be damaging their prospects. Many business users are put off because they associate the term with incomplete or untested software.

It is possible that other Google products will follow the example set by Chrome. The web browser had a beta period of just 100 days and is already at version 2, even though it was only released in September 2008. Could Google’s beta habit be over.

Has Google’s insistence on beta labels ever confused or put-off any of your clients?

Typically, I’d only just installed the latest beta. If that’s anything to go by, Firefox 3.5 is stable and significantly faster than before.

Be aware that many Firefox add-ons may not work yet. However, there is a workaround that forces your favorite extensions to work. Use it at your own risk…

1. Open ‘about:config’.
2. Right-click and add a new boolean value named ‘extensions.checkCompatibility’.
3. Set that value to false.
4. Restart the browser.

The release of Firefox 3.5 is a great topic for my hundredth SitePoint post. Hope you manage to grab the browser whilst it’s hot — let us know what you think of it.

Darren Wood was the front-end HTML/CSS coder chosen for the New Zealand team which completed the winning Rainbow Youth web site. He kindly shares his experiences with SitePoint.

Did you or the team do any preparation in the build up to the event?

Yes. Over the years I’ve amassed a fair few code snippets so I organised these into a quickly-accessible structure. I also downloaded the latest jQuery and all the associated plugins I could need. This was a good thing — the Internet was a little flakey during the competition.

Tim and I had decided to use Ruby on Rails as the back-end technology before the event. I had never used Ruby before, so I did a little research and was pleased to find that RoR is very intuitive and a pleasure to use.

Had you worked with any of the other team members before?

No, but once we started working on the Rainbow Youth website, it felt as if we’d been working together for years. What an amazing team!

You chose BrowserCMS - a relatively unknown Rails-based beta product — as the content management system? What were the key factors that influenced your decision? Did it cause any problems or limitations?

I didn’t have any previous experience with Rails, so Tim’s background was the main factor for choosing BrowserCMS. Overall, there were no major issues with the CMS, despite it being a complex product.

What tools did you use for development? IDE? Source control? Testing suites?

I used TextMate, Photoshop and GIT.

What was the biggest challenge?

My main concern was how I was going to stay awake and alert for 24 hours straight. It turned out to be quite easy. Adrenaline and caffeine take care of most of the heavy lifting. The major challenge was keeping motivated between 4am and 7am.

And what turned out to be easier than you expected?

Working with the team and client. Putting six strangers in a room (well — it felt more like a fish bowl) and asking them to build a fully functional website in 24 hours could have ended badly. Our team and the clients, Clint and Daniel, were fantastic. “A+++ Will trade again!”

Did you need to drop or simplify any features because they couldn’t be achieved within the 24 hours?

Sadly, yes. The client had a huge list of requirements so our first major job was to prioritise. From a front-end code point of view, I would have liked to use more progressive techniques — perhaps HTML5 and sophisticated jQuery goodness.

In which areas do you think you beat your competitors?

It’s tough to say. The Australian team produced an excellent website. I think our strength was the selection of motivational music during the wee hours!

And in which areas did they beat you?

I think the Australian team took more time scoping their project. We attempted to include as many of the client requirements as possible, which probably left our site a little less polished in some areas. They also seemed a lot less stressed than we did!

Would you have done anything differently with the benefit of hindsight?

Tim and I agree that we probably should have rolled our own CMS. BrowserCMS is feature-rich but a little too hardcore to wrangle in 24 hours. I also would liked to have spent more time tweaking the layout of the numerous website forms.

Did you learn anything that you can now use in your day-to-day job?

Yeah — make sure you sleep every twelve hours or so!

Would you enter the FullCodePress event again?

Absolutely! It was an amazing experience which I hope to re-live at some point. Our team was brilliant and we all came away with a solid group of new mates!

Thanks, Darren.

By day, Darren Wood is an interface designer-developer at Marker Studio, Auckland, New Zealand. By night, Darren is known as a web geek and/or “Mr Microformats”. At any time, you can be sure he is an advocate of accessible, semantic HTML.

See also:

• FullCodePress Interview: Tim Connor, New Zealand Team
• FullCodePress Interview: Elle Meredith, Australian Team
• FullCodePress Interview: James Farrell, Australian Team
• FullCodePress: An Interview with Wendy White of Team Australia

Unfortunately, a dangerous precedent is being set by companies surreptitiously installing Firefox add-ons. The worst culprit is Microsoft and the .NET Framework 3.5 Service Pack 1. Most people will receive SP1 as an automatic update, so there is no obvious download or installation. Behind the scenes, the update will install a Firefox add-on named the “Microsoft .NET Framework Assistant”. Microsoft — this looks bad. Very bad…

1. No information
The user is not informed about the add-on prior to, during, or after installation.

2. No authorization
The user can not decline the add-on installation.

3. No uninstallation
The add-on can not be uninstalled via the Firefox Add-on dialog. According to Brad Abrams’ blog, this was a mistake rather than a malicious choice, but it makes you wonder what other mistakes they made in the code?

(Note that Brad’s post links to a patch. Manual removal instructions are also provided, although it involves risky registry tampering.)

4. Additional security risks
The extension enables ClickOnce support. This allows additional software to be installed with minimal user intervention. One of the primary reasons users switch browsers is to avoid the malware issues that plagued IE; how many people want an add-on which circumvents Firefox security?

5. Microsoft is a competitor
At best, this is incompetence. At worst, it’s a serious conflict of interest. Although I do not believe Microsoft intended to sabotage Firefox, this add-on could do anything. Microsoft had the opportunity to make a competing browser slow, unstable, or unreliable — even if that was not their intention.

I suspect this is a case of developer naivety and can only assume the add-on bypassed quality assurance checks because few people were aware of its existence. The company has been working hard to rebuild user trust, but actions like this will not help.

Unfortunately, Microsoft is not the only offender. Take another look at the Add-on dialog above — Sun helpfully installed a “Java Quick Starter” extension with the Java VM. There was no information during installation, it could not be declined, and it can not be uninstalled from the Firefox add-on dialog (an option is hidden deep in the Java Control Panel applet - Advanced > Miscellaneous > Java Quick Starter).

Microsoft and Sun — by all means create Firefox extensions, but there is no need to be unscrupulous. Tell the user, provide opt-outs, or simply release them through the normal Mozilla channels.

See also: What is a Web Browser? No One Knows!

James Farrell was the programmer chosen for the Australian team which completed a great web site for the DDLC NSW. He kindly shares his experiences with SitePoint.

Did you or the team do any preparation in the build up to the event?

TextPattern was our initial first choice of CMS. I spent a lot of time researching the product and installing relevant plugins.

I also familiarised myself with Joomla. That would have been an option had we required a community or multiple user permission handling.

Had you worked with any of the other team members before?

I hadn’t met any of the team until the night before. Our first team meeting was dinner in a Sydney kebab shop.

You chose WordPress as the content management system. That was a safe and solid choice, but what were the key factors that influenced that decision and did it cause any problems or limitations?

It’s not easy to choose a CMS before the event because you have no idea about the requirements. Although TextPattern was our first choice, we couldn’t find plug-ins to fulfil the brief. Joomla was also rejected because none of us had templating experience with the product. We made the late switch to WordPress because most of the team had used it, it has a good range of plugins, and templates are straightforward.

In terms of limitations, perhaps our main problem was that we couldn’t set up a local development environment on our PCs and connect to a single common database server. WordPress saves the URL of the site in the database and, although I’m sure we could have got around this, time was tight and we had to move on. The development environment wasn’t ideal and it probably hindered us.

In the last few hours we were uploading a lot of content and WordPress caused a few issues, such as losing changes. It was a pretty stressful time.

What tools did you use for development? IDE? Source control? Testing suites?

We used Subversion for version control, but didn’t manage our code as well as we would have liked.

What was the biggest challenge?

The main challenge was the unique circumstances under which we were working:

• The event was held during the Cebit Exhibition so random people would poke and chat with us.
• The time constraint meant we were limited in what we could deliver and created additional pressure when problems arose.
• It was frustrating not being able to start work until other aspects were completed. For example, Elle could not code the HTML and CSS until the design had been completed.

And what turned out to be easier than you expected?

Staying awake all night was easier than I expected. Concentration was a problem though — I had just one 15 minute break during the 24 hours. It wasn’t ideal but there’s so much to do in such a short time.

Did you need to drop or simplify any features because they couldn’t be achieved within the 24 hours?

We had to drop our newsletter management system (the Mailpress plugin). It had been working at an early stage but, during integration, it suddenly failed. I spent a while trying to figure out the problem but ran out of time. That was a big disappointment.

In which areas did you beat your competitors and where did they beat you?

I prefer our visual design, but that could be my personal preference or bias. Alexi did a great job with stimulating an emotional response, while communicating a strong positive message.

It’s hard to criticize anything. It’s such a draining experience so I have nothing but respect for what the other team accomplished. It was a competition, but the important thing was that two great charities received fantastic web sites that will revolutionalize their day-to-day operations.

Would you have done anything differently with the benefit of hindsight?

Although WordPress was my favored option, I regret not having prepared more. I spent a lot of time researching plugins on the day.

Ideally, I should have had a few solutions and platforms mapped out before hand. In reality, however, it’s difficult to find the time to prepare when you’re working full-time in the days leading up to the event.

Did you learn anything that you can now use in your day-to-day job?

Definitely. I know WordPress well now and have a few ideas for plugins that I’ll start soon.

Would you enter the FullCodePress event again?

Yes, I would do it again — even though it was the most pressured work situation I’ve experienced and probably took a few years off my life! I met lots of great people, both competitors, organisers and supporters.

Thanks, James.

James Farrell is a freelance web developer based in Sydney. He specialises in high quality, affordable web site design and creative online marketing solutions for small and medium sized businesses. He has been programming since he was 13 and is skilled in web site interface layout, UI design, XHTML, CSS, PHP, Javascript and Flash. You can follow James on Twitter at http://twitter.com/james_farrell.

See also:

• FullCodePress Interview: Tim Connor, New Zealand Team
• FullCodePress Interview: Elle Meredith, Australian Team
• FullCodePress: An Interview with Wendy White of Team Australia
• FullCodePress: WordPress vs Rails
• FullCodePress Update: CodeBlacks Bet on Beta CMS

Five years ago when you talked about application development you were most likely targeting web browsers, desktop computers, game platforms or very limited phone OS’s. On each of those platforms the vast majority of the input was via the user’s fingertips — text, buttons and joysticks.

Today’s devices obviously offer so much more. Touch, GPS, compasses, accelerometers, and proximity detection to name just a few previously unavailable input modes. This has made possible a raft of new types of applications (and I’m not talking about iFart).

Earlier in the year BunsenTech launched a pretty cool iPhone app aimed at the hardcore revhead called Dynolicious. The app uses iPhone’s hardware to track 0-60 acceleration, lateral G force, braking G’s and horsepower amongst other things.

Nifty stuff.

However, now an electric super bike has taken things a step further.

It’s perhaps impressive enough that the very slick-looking MotoCzysz E1pc runs at +120mph on three electric motors with zero emissions.

However, the real kicker for us developers is it uses an iPhone as it’s standard dashboard instrumentation — a bike with a data plan.

There are all sorts of reasons that I think this is super clever.

• Building custom high-tech sensors and screens into a bike is expensive and requires years of testing to perfect. Having them already built-in to a phone that many developers already understand intimately should make development simpler and cheaper. This should be reflected on the ticket price, too.
• The dashboard becomes easily upgradeable and can evolve as MotoCzysz improve and refine it.
• Inevitably geeks, hackers and fans will hack their own dashboards and often come up with stuff way cooler than the default dashboard. You’ve got the possibility of a whole subculture of dashboard hackers evolving.
• I assume you clip the phone out when you leave, so it could easily becomes the equivalent of a key/security device for the bike. In fact, why have a key?

Could we be seeing the beginning of the iPhone (and other advanced mobile devices) becoming part of the default interface/input device for other devices — anything from scooters to ride-on mowers to ski boats to ultralight aircraft?

It sounds like a really interesting niche.

Some of you may have already crawled your way through the innards of Flippa through the numerous beta tests over the last couple of months, but for those who haven’t, let me tell you more about it.

SitePoint’s marketplace has now been divided in two. The entire function of buying and selling web sites has been moved to an entirely separate site, flippa.com. This includes premium, established, and start-up web site sales.

So if you’re in the market to buy or sell a web site, then flippa.com is your new home. There’s a whole raft of new features to go with the new name, but I’ll let the Flippa team tell you all about that.

If you’ve got a SitePoint Forums or SitePoint Marketplace login, then your credentials will allow you to log in to Flippa. Additionally, your complete auction history, bidding history and feedback rating has also been carried over to the new site, so all you need to do is log in and you’re good to go!

Oh, and if you’re new to buying and selling web sites, welcome to the jungle! You can sign up here.

So what’s left in the SitePoint Marketplace?

Our vibrant Looking To Hire, and Advertise Your Services marketplaces will remain on SitePoint, as will classified listings for scripts, hosting deals, content for sale, advertising space for sale, and more. So if you’ve got something to sell (that’s not a domain or web site), or you’re seeking that next gun recruit, look no further than the SitePoint Marketplace!

With all web site buying and selling now conducted on flippa.com. The SitePoint Marketplace will be more focused on advertising your services and general webmaster classifieds and we’re looking for your suggestions on where to from here.

Some ideas that we’ve discussed include:

• Expanding the Looking To Hire and Advertise Your Services sections of the marketplace to include specific areas for designers, programmers, writers & bloggers, and marketers & SEO Gurus,
• The introduction of web sites reviews including SEO, usability and design feedback

That’s just a couple of possibilities to get you started.

To open up the flow of ideas, we’ve created a form thread to capture your thoughts. Tell us your vision for the SitePoint Marketplace. What do you think the SitePoint Marketplace should focus on?

Note, any feedback about flippa.com is best directed at the Flippa team — you can talk to them on Twitter, make suggestions on their UserVoice page, or lodge support tickets via their ZenDesk forum.

The Millionth English Word

According to Global Language Monitor (GLM), at 10:21 GMT on 10 June 2009 the English language consisted of precisely 999,999 different words. The millionth word appeared 1 minute later and earned its place in the dictionary. That word was: “Web 2.0″.

Despite “Web 2.0″ being an existing word with a digit on the end, GLM received a significant amount of publicity for their announcement. They are a commercial company who scour the Internet for new terms. A word must be used at least 25,000 times across national boundaries and outside specialisms for it to pass GLM’s rigorous validation procedures.

My analysis may not be as thorough, but a simple search on Google reports over 45 million references. Yahoo reports 173 million references. Bing does even better with 2.2 billion references (so much for it being a “decision” engine!). It’s strange that Web 2.0 had not been considered before; the term has been around almost 10 years and has certainly been popular since 2004.

Lexicographers were quick to point out that the exact size of the English vocabulary is impossible to quantify. The Oxford English Dictionary also estimates that the number of words is nearer two-thirds of a million.

Whether it’s hype or a ridiculous publicity campaign, I’m looking forward to “Web 2.0″ appearing in the dictionary. Perhaps we’ll finally have a description of what it really means!

The Correct Pronunciation of “GIF”

Graphics Interchange Format, or GIF, was introduced by CompuServe in 1987. Show someone the word “GIF” and ask them to read it aloud. Most English speakers will pronounce it with a hard “g” as in “gift”. Unfortunately, they are wrong — it’s pronounced “jiff”, with a soft “g”, and the evidence is presented at the GIF Pronunciation Page.

Personally, I don’t think it matters what the original programmers intended; almost everyone pronounces GIF the way it is written. There are pedants who insist on using “jiff” but no one understands them! Do you prefer GIF or JIFF? I’m firmly in the GIF camp and my campaign to correct its pronunciation starts here!

(Incidentally, the PNG specification also insists on “ping” rather than “pee en gee”, but it’s an open standard so we can change that!)

According to a leaked communication from the company:

To ensure that Microsoft is in compliance with European law, Microsoft will be releasing a separate version of Windows 7 for distribution in Europe that will not include Windows Internet Explorer.

Microsoft will offer IE8 separately and free of charge and will make it easy and convenient for PC manufacturers to preinstall IE 8 on Windows 7 machines in Europe if they so choose. PC manufacturers may choose to install an alternative browser instead of IE 8, and has always been the case, they may install multiple browsers if they wish.

Two flavors of each Windows 7 (Home, Pro, Ultimate, etc.) will be available in EU territories:

• Windows 7 E: the release without Internet Explorer, and
• Windows 7 N: the release without IE and Windows Media Player.

There are several reasons why this is a clever move by the software giant:

1. Microsoft are complying with EU law
Windows 7 E/N will have no default browser so Microsoft appears to be committed to upholding competition laws. I’m no legal expert, but I’m sure the EU will have a tough time proceeding with the anti-trust case.

2. IE is still available in XP and Vista
XP and Vista users already have IE installed — will upgrading to Windows 7 remove the browser? It seems unlikely and nor would anyone want that to happen.

3. IE is only hidden
Windows depends on browser components for several core services so, although IE can be uninstalled in Windows 7, it only affects the OS at a superficial level.

Windows 7 E/N is likely to provide IE in an ‘uninstalled’ state. What would you do following a clean installation of the OS? Hunt around for browser installers or quickly re-enable IE and download what you need?

4. OEMs will still choose IE
Microsoft has made it clear that PC manufacturers can offer an alternative or a choice of browsers. How many OEMs will do that?

• A browser choice screen requires a custom application and up-to-date installers. Developing and maintaining that software will have a direct cost to the manufacturer.
• Alternatively, assume that an OEM offered a different browser. Would novices be confused by the absence of the blue ‘e’ icon and complain that their new PC is not Internet-enabled?

Anyone who uses Firefox, Opera, Safari, or Chrome already downloads and maintains their browser of choice so a default IE installation is unlikely confuse them. That is not necessarily true for IE users. Providing an alternative browser will increase the OEM’s support costs and they cannot pass people directly to the Microsoft helpline.

5. Microsoft will not need to promote other browsers
The EU and Opera’s preferred solution is a ‘choose a browser’ screen shown during Windows installation and possibly within the Control Panel. That would have been a logistical nightmare for Microsoft and I’m sure vendors would still have complained their browser was shown less prominently than others.

The solution also required Microsoft to publicise other browsers. One of the reasons IE’s market share remains high is because users are not aware of alternatives or are unable to install them.

Microsoft’s decision to remove Internet Explorer from the European version of Windows 7 will have an negligible effect on the browser’s market share. That may not be the outcome Opera or the EU wanted, but I suspect the anti-trust battle is at least 10 years too late.

With Opera Unite, developers will create applications — or Opera Unite Services — which run in the end-user’s browser, connect directly to other computers, and share data. Sample scenarios outlined by Eng include media sharing, instant communication, and anything in a “class of social software on the web.” The end goal suggested in Eng’s announcement includes restoring power to end-users to communicate with each other, using their own infrastructure (personal computers, residential internet connections etc.).

In an article on Dev.Opera, Chris Mills explains how to get up and running with the latest Opera Unite build. Mills offers a fairly concise description for Opera Unite — “a collaborative technology that uses a compact server inside [Opera] to share data and services.”

As Google, Yahoo! and Microsoft seek to create centralized platforms and capture mindshare, Opera Unite aims to steer the web away from a centrally-managed architecture run by “computers with more power than the rest” (servers). Opera believes that a new set of opportunities exists where servers are not required and applications can communicate directly with peer systems. Should Opera succeed, this would strike at the heart of user-driven content sites, many of which rely on users sharing media with a select group of friends for traffic. Interestingly, Opera Unite seems to make no attempt to challenge fundamentally centralised systems, such as search engines, nor does it attempt to replace online communities that thrive on sharing content to a general audience (YouTube, anyone?).

A quick run through of Opera Unite is promising — a wizard in the Opera 10 beta (available under the Tools menu) takes the user through the process of creating an account, and helpfully offers to attempt to configure UPnP (no port forwarding instructions in sight — the system is clearly optimized for as little configuration as possible). Once set up, the Opera Unite sidebar appears, and a list of default services is available.

Enabling the file sharing application, for example, adds a File Sharing entry to a list of available services on a public URL (generally http://(computer name).(Opera username).operaunite.com). This public URL represents a landing page of sorts for the user, and Opera suggests this is served from the end-user’s machine, although the hostname of the public URL resolves to an Opera server. Additionally, while any browser can be used to browse available content, it appears that Opera Unite is required to actually access content, e.g. download shared files.

The initial release feels extremely stable, although Opera emphasizes that the Unite technology is still in alpha, and there will undoubtedly be challenges with complex networking environments (NAT, no UPnP etc.) for which the technology is yet to be battle-tested. Builds are available for Windows, Linux and Mac OS X.

Download Opera Unite and let us know what you think.