Archive for December, 2009

File Under: Browsers
Dec
18
2009

Firefox 3.6 Nearly Complete, Fifth Beta Available Now

By

Mozilla has released the fifth, and reportedly last, beta version of the next Firefox browser.

Firefox 3.6 has seen more than 100 bugfixes since beta 4, including improvements in Firefox’s performance, stability, and security. If you’d like to take beta 5 for a spin, head over to the Mozilla downloads page or just wait for your current Firefox 3.6 beta to automatically update.

Five betas is an unusually high number for a Firefox release, but the latest package clears most of the remaining blocking bugs. This means the final release of Firefox 3.6 is likely just around the corner.

Given that we’re so close to the final release, Firefox 3.6 beta 5 doesn’t contain any significant new features. It does, however, bring some very welcome bug fixes, including a couple of annoying scrolling bugs we’ve had to deal with in previous beta releases.

This release also fixes a bug that would occasionally cause Firefox to crash on the Twitter login page, and worse, would expose passwords in plain text.

The other big of good news for this release is that Add-on developers are getting ready for the final release of Firefox 3.6. Mozilla reports that over 70 percent of Firefox Add-ons have now been upgraded to be compatible with Firefox 3.6 Beta, including the ever-popular Greasemonkey.

Sadly two of our favorite web development add-ons, Yahoo’s YSlow and Google’s recently released Page Speed tool have yet to update for Firefox 3.6. If you’d like to help out the developers of your favorite add-on, grab the Add-on Compatibility Reporter, which, among other things, allows you to run add-ons that haven’t yet been updated.

See Also:

Tags: , ,
File Under: HTML5, Mobile, Web Apps
Dec
17
2009

Google: Mobile Gmail Exemplifies HTML5’s Power

By

Over the last few months, Google has been busy optimizing its free Gmail service for mobile phones.

In a post on the Google Mobile blog, the Gmail team highlights the various optimizations it has steadily been building into the mobile version of the webmail site. The result is a Gmail web app that exhibits many of the same behaviors one would expect from a native client application, and a site that now loads, according to Google, two to three times faster than it did eight months ago in April, 2009.

The biggest shift towards a faster, more sexy Gmail has been the proliferation of modern web browsers which support HTML5′s many advancements, according to Google engineering manager Alex Nicolaou, the author of Thursday’s post.

Mobile Safari, the Android browser, Firefox Mobile and Opera Mobile are thoroughly modern browsers that embrace the emerging HTML5 standard, and the fact that more people are using them on their advanced phones means that Google can serve out faster, more app-like versions of its web apps.

“So yes, HTML5 and the mobile web are clearly up to the task of building rich and powerful apps,” Nicolaou writes.

His comments come during what has been a week of intense debate among developers over which platform they should choose for their apps — native iPhone and Android apps, or the web.

Of course, the web has the greatest reach since it means you can gain users on every platform where a browser exists. Web apps have other advantages, but you sacrifice some key things when you go the web route — a richer user interface, advanced scrolling behaviors and animations, specialized layouts and fonts, offline access and interactions with hardware sensors on the phone.

The great promise of HTML5 is that it will do away with those limitations, giving developers the ability to put new types of advanced interactions into their browser-based apps.

HTML5′s arrival doesn’t quite signal the end of the downloadable, native application, but it does blur the line between the platform-based device operating system and the broader web operating system.

Google recognizes that as well.

“It’s also worth noting that as a worldwide mobile team, we’ll continue to build native apps where it makes sense,” Nicolau says. “But we’re incredibly optimistic about the future of the mobile web — both for developers and for the users we serve.”

See Also:

Tags: , ,
File Under: APIs, Web Services
Dec
17
2009

Craigslist Reverses Yahoo Pipes Ban, But Developers Have Already Moved On

By

Earlier this month, Craigslist blocked Yahoo Pipes from accessing any Craigslist page. As a result, mashups all over the web were suddenly without data.

After not responding to inquires, Craigslist CEO Jim Buckmaster eventually posted a terse message on the company’s blog saying that Yahoo Pipes mashups were using “a disproportionate amount of server/bandwidth resources.”Now it would seem that Craigslist has ended its ban of Yahoo Pipes, but for developers we imagine the damage has been done. For its part, the startup Flippity, which was the first to notice its Yahoo Pipe had been blocked, says it has moved on to friendlier sources, rewriting Flippity to use the eBay API.

It wasn’t the first time Craigslist has shutdown outsiders trying to improve on the site’s famously antiquated tools — the site previous blocked ListPic, a tool designed to help Craigslist users browse by images, and a tool to search all the Craigslist sites at once.

As we said in our initial report on this debacle, if you’re a developer looking for data to use in a mashup, think twice about Craigslist. The site has a wealth of data, but it guards it jealously and has no qualms about blocking even major players like Yahoo.

See Also:

Tags: , , ,
File Under: Browsers
Dec
17
2009

Personas 2.0 Makes it Even Easier to Customize Firefox

By

Mozilla has bumped Personas to version 2.0. The company has upgraded its set of lightweight themes for Firefox in anticipation of Firefox 3.6, which will be the first version of the browser to support Personas out of the box. Other versions of Firefox, like the current Firefox 3.5, still require that you download and install the Personas add-on to manage your themes.

While Personas 2.0 offers a few new features — like an icon in the status bar that allows you to quickly switch themes — the primary focus has been on redesigning the Personas gallery and making the installation process smoother and easier.

The gallery has a host of new ways to discover Personas, and browsing through it is now much more like browsing the halls of the Firefox add-ons gallery. New means of filtering and sorting themes by category, popularity or related Personas make finding what you want a bit easier.

If you’re syncing Personas through the Weave add-on, there’s no need to worry; we were able to sync 2.0 Personas through Weave with no difficulty.

Personally I’ve never understood themes, for Firefox or anything else. Themes are a bit like putting stickers on hammer — You can, but why spend the time? I like my browser, but I like because its useful, not because it has pictures of Harry Potter in the menu bar. However, according to Mozilla, the Personas site will likely see its 40,000th design uploaded by the end of 2009, so clearly at least some of you are enjoying Personas.

While Personas aren’t likely to convince hesitant users to switch to Firefox, they do offer an easy way to make your browser a bit more, well, personal. If you’re a fan, head on over the revamped Personas site and be sure to install the Personas add-on if you aren’t using the Firefox 3.6 beta.

See Also:

Tags: , ,
File Under: Identity, Social, UI/UX
Dec
17
2009

OpenID: Over One Billion (Potentially) Served

By

OpenID, the single sign-on solution which allows you to use a unified identity across the web, now boasts one billion potential users. Providers like Google, Yahoo and WordPress have adopted the technology, providing nearly everyone on the web with easy access to an OpenID account.

OpenID lets you log in to your favorite website using only your e-mail address or a URL — your blog’s address, a profile page on a social network or your social network username/password. Using one of those identifiers, you can log in to any website or service where OpenID is welcome, saving you the trouble of having to keep track of dozens of account names and passwords. There are also companion technologies that help you automatically fill out a profile and connect you with your friends once you’re logged in to a new social website.

For a long time, OpenID was a fringe technology, and few large players supported it. In January 2008, Yahoo and AOL were the first major destination sites to host OpenID accounts. 2009 has seen everyone from Microsoft to Facebook to the U.S. Government embracing OpenID. In addition to the one billion accounts coming from OpenID providers, the OpenID foundation says that nearly 9 million websites will allow you to login using your OpenID credentials.

The short story is that OpenID is now well established on the web. But the story doesn’t end there.

Sadly, one billion potential users does not one billion users make. Many people with OpenID accounts remain blissfully unaware of OpenID and what it can do for them. OpenID also faces strong competition from proprietary ID solutions like those of Facebook or Twitter.

OpenID interfaces are another problem we’ve covered before — different sites use vastly different sign-in forms which has creates confusion for less-than-savvy web users. Couple that with Facebook’s far simpler Facebook Connect tools and you begin to see why OpenID doesn’t have one billion actual users.

The good news is that the OpenID Foundation and its partners have been working hard to streamline the login process and improve the usability of OpenID on those 9 million sites that accept OpenID.

We’re excited to see that what began as little more than a grassroots effort to solve the problem of remembering too many usernames and passwords, has turned into a massive, web-wide effort to create better, portable identity tools. So even if OpenID hasn’t seen the widespread adoption of other login systems, it certainly set the ball rolling among the web’s social networking technicians.

See Also:

Tags: , ,
File Under: Identity, Social
Dec
16
2009

Google Crawlers Now Understand ‘Canonical’ URLs

By

Migrating a web site from one domain to another is never easy. You’ll probably lose whatever Google ranking your old pages had, possibly break incoming links and generally disrupt the flux capacitor of the web.

Of course, there are occasionally good reasons to move your content and now there are some new ways to let Google know what you’re up to. The Google Webmaster blog recently announced that Google will support the cross-domain rel="canonical" link element. That means you can effectively migrate your site to a new domain even if you don’t have server access to do redirects.

In most cases, Google still suggests that, if possible, you use 301 permanent redirects to point both visitors and search engine bots to your new domain. However, if that’s not possible for some reason, (for example, if you’re migrating from a hosted blog service to your own domain) then you can add rel="canonical" element to your page headers and Google will index the new URL.

Note that in our example — moving from a hosted blogging service to a self-hosted domain — it’s OK if there are some differences between the new and old pages, but the basic content (the blog post) should be the same.

Previously, Google would look down on cases of duplicate content across domains. Given the number of content-stealing “splogs” out there, filtering duplicate content by domains is a good way for Google to stop search engine spam. The problem is there are legitimate reasons to have duplicate content, like migrating a site to a new domain, and now there’s a way to do it.

One important note, Google no longer recommends blocking access to duplicate content on your website, whether with a robots.txt file or other methods. Just use the rel="canonical" tag instead.

See Also:

Tags: , ,
File Under: Web Apps
Dec
15
2009

This Is the Guy Behind Reddit’s Secret Santa Gift Exchange

By

Webmonkey got the chance to sit down with Dan McComas (aka kickme444), creator of the amazing Reddit Secret Santa program, for an exclusive interview. Since launching redditgifts.com a month ago on November 16, around 5,000 participants from the Reddit community have taken part in what is believed to be the largest Secret Santa program ever.

We spoke with Dan, who by day is a systems architect from Alameda, California (and a camera shy one at that, hence the portrait), and got some details about how the project came to be, how he built and launched the website in such a short amount of time, and whether he’s going to do it again next year.

The Story:

Dan thought up the idea on November 10. He posted on Ask Reddit wondering aloud if anyone would want to participate in an anonymous gift exchange.

Ideas began flooding in, so Dan created a Secret Santa subreddit and encouraged other users to subscribe to it. Within a couple of hours, over 1,000 users had subscribed and people began volunteering for the various technical tasks (setting up the servers, programming) involved with launching the project.

The community quickly fleshed out a set of rules — you had to have signed up as a Reddit user before November 10, for one, and they set a $15 cap on gifts.

Dan realized something this big couldn’t be administered by hand, so he created a quick site in Django and parked it at RedditGifts.com. At first, all it did was provide information and the rules and take sign-ups through a form.

The site launched on November 16th, less than a week after the idea was proposed. By this time, a small army of developers had come together to help make the project a reality.

The back-end was fully automated, so all Dan and his wife had to do was provide customer service via e-mail. In total, they sent out over 30,000 e-mails — an aggregate of different bulk e-mails to different subsets of users who each had different issues or needs. But still, lots of e-mails.

Around 5,000 people signed up to exchange gifts. By the night of Tuesday, December 15th, all but 300-500 had followed through. That’s a success rate of around 90 percent.

Reddit users spent over $114,500 on gifts. Yes, many people broke the $15 rule. The real gift was to FedEx and UPS — over $47,000 was spent on shipping. View the full stats here.

Reddit users were matched up with strangers, so they had to casually stalk one another on Reddit to learn as much as they could about their recipients’ personalities.

During sign-up, people were given a blank text field for special requests. Most left it blank, but others put book titles, food allergies or other special requests.

Most gave one gift and received one gift, but some people gave two or three gifts.

Participants were encouraged to document the gifts they received, and you can see the various unboxing shots on a Gift Gallery. In true Reddit fashion, you can vote up the cooler gifts.

Dan was matched with a 17-year-old from San Diego (his only request was that he received something his parents wouldn’t freak out about) and an avid Reddit user who was a bit older. Dan walked around the shops in his home town of Alameda and put together some packages of novelties like Space Invaders ice cube trays and sci-fi memorabilia. The avid Reddit user got a rare Upvote Breakfast poster.

Dan received two gifts. He got an ancient wooden ring from Ireland from one person, and a Gameboy Advance from somebody else — either from Sweden or Finland, he’s not sure. It came with a Donkey Kong game and a European power plug, so he can’t use it without an adapter.

Will he do it next year? He’s not sure. He and his wife are exhausted from all the work. Much of the informational copy, the sign-up form, the database and the rest of site he built are all reusable, but this project pulled in more traffic than any website he’s ever created during his professional career.

Wednesday December 16th will mark one month since the site’s launch. All that in a month! Amazing.

Tags:
File Under: CSS, Visual Design, Web Standards
Dec
15
2009

Slick Web Design Gets Easier Thanks to CSS 3’s Transform Tools

By

Now that modern browsers are stepping up with better support for CSS 3, web designers have even more powerful transforms to play with. Page elements can be animated, rotated or otherwise transformed in ways that previously weren’t possible, or at least required complex JavaScript.

Of course, browser support is still somewhat limited in “the real world,” so it’ll be some time before you can begin using them heavily in your daily work. But if you’d like to jumpstart your understanding of CSS 3, the 24 Ways blog has a nice tutorial on manipulating images using CSS 3′s transform properties.

The article tackles mainly WebKit-based transforms, though if you’re using nightly builds of Firefox 3.7, the -moz-transform code will work as well.

Obviously, Internet Explorer won’t be able to render any of these CSS rules, so you’ll want to make sure the effects degrade for browsers that don’t understand CSS 3. For example, the first demo in the tutorial simply rotates an image and applies a drop shadow. Browsers that don’t understand the CSS would simply display the image without the rotation or shadow — not ideal, but workable.

The other possibility would be to include some JavaScript for less-capable browsers, but that defeats much of the purpose of CSS 3 — making complex transforms quick and simple.

Another point worth mentioning is that the box-shadow rule has actually been dropped from the CSS 3 spec due to time constraints. Most likely it will be resurrected as drop-shadow, or something similar, but the change is still in the discussion stage so you might want to hold off using box-shadow for now.

Be sure to dig around the rest of the 24 Ways site, which is a bit like an advent calendar for web designers. We’ve featured 24 Ways tutorials in the past and, so far, this year’s offering look to be every bit as useful as those from years past.

See Also:

Tags: ,
File Under: Web Services
Dec
15
2009

Google Jumps on the URL-Shortening Bandwagon With Goo.gl

By

Given the rapid proliferation of URL-shortening services of late, it was only a matter of time before Google got into the game.

Now the company has debuted its own URL-shortening service at the new domain goo.gl. (That’s the top-level domain of Greenland, by the way; its awesome coat of arms is at right).

At launch, goo.gl integrates with Google’s FeedBurner publishing tool and Google Toolbar, but the company plans to roll it out as a more full-featured web service later. At least for now, outside apps like Twitter clients won’t be able to access the goo.gl domain to shorten links, which means there’s still a place for Bit.ly and the other services like it. But with Facebook also rumored to be working on a homegrown URL-shortening service, and with Bit.ly now offering a white-label “pro” version of its service that would give any site the ability to run its own custom URL shortener, we suspect the days of dedicated URL shorteners are numbered.

The FeedBurner integration allows FeedBurner customers to automatically send updates from their RSS feeds to Twitter and any other services where limited post lengths dictate the need for shorter URLs.

That’s bad news for a third-party service like Twitterfeed, which has long enabled bloggers to automatically tweet new posts using their RSS feeds. If you’re already using FeedBurner to host and track your site’s RSS feeds, now there’s no need to rely on outside services to post to Twitter. The new FeedBurner option also takes advantage of the Twitter’s OAuth support, so you won’t need to hand over your Twitter username and password to Google.

You can also use the new goo.gl service through the Google Toolbar, which will allow you to post shortened links to any page, much like you would with Bit.ly and other services.

Right now, Bit.ly is the URL shortener favored by Twitter, and is therefore the largest such service. Bit.ly collects massive amounts of data on the billions of clicks that flow through its short URL service every month. Such data mining is extremely valuable for anyone in the businesses of advertising and real-time search, so we can expect Google to be keeping a close eye on every click at goo.gl. The privacy policy for the goo.gl service says, “Google may choose to publicly display aggregate and non-personally identifiable statistics about particular shortened links, such as the number of end user clicks.” If you don’t want Google harvesting this data, then the new service is not for you.

Of course the bigger problems with URL shorteners — link-rot, spam and redirect mishaps to name a few — are still problems regardless of whether the shortener is controlled by Google or anyone else. And for anyone who thinks that Google services have a better chance of being around far into the future, may we remind you that Google Notebook, Google Answers and several other services have disappeared over the years.

As always, if you really want a future-proof URL shortener, your best bet is to build your own using one of the many libraries and apps available for Python, PHP, Ruby and other popular programming languages.

See Also:

Tags: ,
File Under: Web Apps, Web Services
Dec
14
2009

Use Google for Text-to-Speech Translations in the Browser

By

Google Translate has rolled a out several new features, and one of them lets you hear your translated English text spoken by a computerized voice.

For maximum cross-browser compatibility, Google Translate embeds text-to-speech mp3s using Flash. However, web developer Weston Ruter has already figured out how to grab the audio files for use on any webpage and uses the HTML5 audio tag to embed them in his demo.

The new Translate audio tools generate (rough) pronunciations of whatever text is entered the end of this URL:


http://translate.google.com/translate_tts?tl=en&q=text

Just pass along the phrase you’d like to hear instead of “text” at the end of the URL and you’ll get an mp3 file like this one. You may have to try a couple of different browsers to get a translated file that works flawlessly. (It works especially well in Chrome, which is not surprising.) With a tool like this, the urge to get silly is irresistible, as Twitter user Brad Cohen has done by feeding Google Translate some Justin Timberlake lyrics.

Take one of these mp3s and embed it using HTML5′s audio tag — as Ruter has done on his site — and you’ve got an instant, mobile translation service that’ll work on the iPhone, Android and other Webkit-based mobile browsers.

The iPhone handles Ruter’s example quite well, though the interface is slightly awkward — Mobile Safari brings up a new overlay page to play the audio. Still, despite a few quirks, having a text-to-speech generator available on the web offers developers quite a few browser-based possibilities for a task previously relegated to desktop apps, or, otherwise, full-fledged mobile apps like the translation tools available for the iPhone and Android devices.

The biggest catch at the moment is that Google Translate’s text-to-speech feature is only available in English. But as with most Google services, the initial limitations are likely to disappear before too long.

Also part of the announcement, Google Translate now performs translations as you type — no need for the “translate” button — and there’s also a new “Show romanization” that lets you read the text written phonetically in English. This works with all non-Roman languages except for Hebrew, Arabic and Persian.

See Also:

Tags: ,
« OLDER POSTS