Archive for November, 2009

File Under: Location, Mobile, Web Apps

Get Your Fix of Hyper-Local News With Fwix

Readers jonesing for local news on the web now have one more place to turn.

News website Fwix has emerged a strong contender in the quickly growing field of hyper-local news aggregators, sites which filter local and area news stories for you, showing you the hottest links in your ZIP code.

The site aggregates stories and videos from about 100 cities worldwide, mostly in North America. Each city has its own home page, and each city’s page has different sections for politics, crime, sports and other topics.

Local news headlines are presented with the newest stories at the top, so you get a constantly-updating river of news about topics you care about in the area where you live. New stories flow in as they become hot, appearing dynamically without a page refresh.

It’s a design of the times. Twitter and Facebook have turned us into breaking news junkies. Of all the hottest stories being passed around on the real-time web, few are more interesting than the ones happening closest to us, whether in our city or in our immediate neighborhood. And geoloaction tools in our computers, phones and even in our web browsers are making local content easier to find. To that end, we’ve seen a number of hyper-local news aggregators come down the pike. Sites like EveryBlock and Outside.in offer hyper-local experiences complete with maps, and larger sites like Google News give us the ability to filter results to state, county or city level.

Much like other aggregators, Fwix pulls content from a handful of mainstream news sources, like major area newspapers and local TV network affiliates, as well as social media sites and dozens of local blogs. Check out its long list of sources for the San Francisco bay area.

According to Fwix founder and CEO Darian Shiraz, who gave us a hands-on tour of the site, news sources are hand-picked by Fwix’s small editorial team. Users can also request the addition of a news source by submitting a URL on the Fwix website.

The free Fwix iPhone app, pictured at the top, adds another way in. It has a “Report News” button — click it and you can submit a photo of something happening in your city. So if you’re walking around and you see a building on fire or a gnarly car crash, take a photo with your iPhone and post it (after you call 911, of course).

For all of its depth, the year-old service isn’t yet perfect.

News headlines on the Fwix front page are chosen based on popularity, which is determined by algorithms written by Fwix (using disco project code). Popularity is largely measured by how many clicks a story has gotten and how many sources are reporting an event.

The results can get wonky. When San Francisco Giants pitcher Tim Lincecum was busted for pot possession, there were three headlines about it at the top of the SF bay area page. All were from different sources, but since most news outlets just ran a wire story, each link had the same headline, the same summary and the same information. Likewise when Lincecum won the Cy Young award — everyone had the story, so the first three links on Fwix were all of the same.

Other hyper-local services have the same problem, but Google News deals with the issue rather elegantly by nesting headlines for related stories.

Also, Fwix isn’t yet a reliable source of breaking news. The day the Bay Bridge was shut down due to structural damage — a huge news event in San Francisco — there was nothing about it on the Fwix homepage even hours after it happened. Also, the bridge closure presented a perfect opportunity for users to submit relevant photos and tweets, but I didn’t see any show up on the website.

Still, while its method of presenting aggregated news may be more chaotic than what Google News offers, Fwix does surface a wider range of stories. Every time I’ve visited the Fwix home page or fired up the iPhone app since I began regularly using both a month ago, I’ve found at least one or two interesting headlines I never would have encountered had I been left to my own devices. Several times, it was something in my neighborhood that one of my favorite go-to local blogs had overlooked. By regularly visiting the site, I’ve discovered some great local blogs I hadn’t heard about.

Also, interesting videos show up often — something few other hyper-local news sites can claim.

While content providers and the large-scale aggregators continue to butt heads over who has the right to profit from news content, Fwix, relatively speaking, takes the high road.

The company has recently launched its AdWire service, an ad network that splits revenue with the site owners running the ads and the publishers of the stories they’re linking to, many of whom are hobbyists, citizen journalists or freelance bloggers.

Along the same lines, Fwix only posts very short, one or two-sentence summaries of news stories, then offers a clearly-labeled link to the original source, much like an RSS reader. The original story’s page is wrapped in a frame with buttons to share the link on social sites using Fwix’s own URL-shortening service. Click-through content frames are a no-no in our book, but at least this one is easy enough to dismiss. Just click the giant “X” button.

See also:

File Under: Identity, Social

Google Profiles Now Function as OpenIDs

So says Google’s Brad Fitzpatrick in a Twitter post Wednesday.

You can now use the URL of your Google Profile to confirm your identity on any website that supports OpenID. When the site asks you for an OpenID identifier, just plug in the URL of your Google Profile and you’ll be directed to Google, where you confirm the request.

OpenID Foundation board member Chris Messina has posted a screenshot of what the user flow looks like when using your Google Profile URL to log in on a website that supports OpenID:

Brad is one of the creators of OpenID and one of the driving forces behind Google Profiles. Google launched its public profile service, which allows anyone with a Google account to create a public profile on the web that shows up in Google’s search results, earlier this year. At first, Profiles were rather spare, but Google has slowly been enhancing the features of Profiles to include vanity URLs and support for microformats.

These profiles are advantageous over proprietary social networking profiles because of their high visibility in Google, the depth they allow, and because they function as a social hub — most people use them to point to their social presences on other sites. Not to mention that Google Profiles appear on the open web rather than inside of Facebook, where, by default, a profile can only be seen by people you’ve connected with on Facebook.

Webfinger, also referenced in Fitzpatrick’s tweet, is a new protocol Google is building into Gmail. It lets you attach any public identity data to your e-mail address. Learn more about it at the Google Code project site.

See Also:

File Under: UI/UX, Web Services

Google Tests Redesigned Search Page

Google’s new look? The search giant is testing a revamped results page. Click the image for a larger view.

Google appears to be testing a possible redesign of its iconic search page. Whether or not the new prototype will ever become official remains unknown, but thanks to some clever JavaScript you can check out the new look today.

The Google watchers over at Google Blogoscoped have found a snippet of JavaScript you can paste into your browser’s URL field which will activate the new look. Because the JavaScript code sets a new cookie, you’ll most likely need to log out of your Google account before it works.

Once the cookie is set, refresh the Google homepage and you’ll see the changes. The search buttons have become blue and the overall look is a bit like that of Google Wave. More significant is the redesigned search results page (seen above) which features an always-on sidebar for narrowing search results by type, date and view.

The brighter, more Wave-like look of the prototype doesn’t bother us, but we’re not so sure about the sidebar, especially given that the same options are already available in the infinitely more compact menu that runs along the top of the page.

There is one new search option in the sidebar that you won’t find on the current Google page — the ability to see results from online forum sites.

The good news, should the new look utterly disgust you, is that so far Google hasn’t even mentioned the new look (and had not responded to our inquires when this story was published) let alone taken any steps toward making it official. Given Google’s track record of beta testing, we suspect the redesign will be thoroughly and publicly tested before it goes live, if in fact it ever does.

See Also:

File Under: Browsers, Software

Google Lays the Groundwork for Extensions in Chrome

Google is getting ready to offer widespread support for extensions in Chrome, launching a program which will allow third-party developers to add features to its browser.

The company released more details about its new Chrome Extension Gallery Tuesday. Developers can now upload their extensions to the Chrome Extensions Gallery, in effect publishing their extensions even before the browser officially supports them. Support for Chrome extensions is available in the current developer release, but they will probably arrive in the browser for all users before the end of the year.

At the moment, there isn’t much to see in the Chrome gallery — it’s just a form for developers to upload their extensions. However, Google is clearly hoping Chrome will one day support an extension ecosystem similar to the one Mozilla enjoys with its highly successful add-ons community for Firefox. The site will offer the ability to search and browse for extensions, and Google is encouraging developers to upload videos and screenshots explaining what each extension does.

The company has also posted guidelines outlawing things like copyright infringement, hate speech and any extension to “enable the unauthorized download of streaming content or media” — which means we probably won’t see extensions for ripping videos from YouTube.

To that point, Chrome extensions will be reviewed before they become publicly available. The Chrome Blog says that, for most extensions, “the review process is fully automated.” However, if your extension plans to use low-level components or access file:// URIs, Google will require a manual review and may ask developers for additional information before such extensions end up in the gallery.

See Also:

File Under: Browsers, Web Services

Latest Opera Update Hopes to ‘Unite’ the Web

Opera software has released an update to its flagship desktop web browser. This update incorporates Opera’s lightweight web server Unite, previously only available as a separate beta download, into the browser.

Opera 10.10 is an incremental update to Opera 10, released in September. If you’d like to take the new browser for a spin, head on over the to official site and download it. Opera 10.10 is a free download for all major platforms.

The only new notable feature in this release is Opera Unite, which is essentially a webserver built in to the web browser. Opera Unite allows users to build and host not only web sites, but also custom web apps powered by JavaScript. Apps running on Unite could be used to power private social networks like mini-Facebooks or mini-Flickrs, as well as collaborative tools like Google Wave or even file-sharing darknets.

With Unite, Opera is making it possible for anyone to host their own server with the click of a button. Once Unite is up and running, anyone can connect to it from any browser. While your copy of Opera runs the server, your visitors can use any browser they’d like.

Opera Unite is probably the easiest way to self-host a web site that we’ve seen. However, it’s not without some significant drawbacks — namely the issues of bandwidth and uptime, which we covered in detail in our initial review of Opera Unite.

While upload speeds from most ISPs are reasonable, don’t expect your Opera Unite pages to perform like a server sitting at the end of fiber optic cable. Then there’s the uptime problem… put your laptop to sleep and your locally-hosted website disappears from the web.

Still, for casual purposes, like hosting some photos for friends and family or sharing some files, Opera Unite works admirably well.

See Also:

File Under: HTML, Web Apps

Parsing HTML? There’s an App for That

Parsing HTML and extracting useful bits of data is common job for web programmers. The only question is, how do you go about it?

You could use Regular Expressions. On the surface of things, regular expressions seem like a natural fit — after all, HTML is just strings of text right? What better way to parse strings of text than with regular expressions?

Well, as anyone who’s tried using regular expressions to parse HTML can tell you, there be dragons.

If the HTML you’re parsing is well-formed and syntactically correct, using regular expressions actually isn’t so bad. Sadly, this is rarely the case. Given the level of cruft in the average chunk of HTML, parsing such strings with regular expressions is enough to drive even the most mild mannered programmer to the brink of insanity, as witnessed by this humorous passage on Stack Overflow:

Every time you attempt to parse HTML with regular expressions, the unholy child weeps the blood of virgins, and Russian hackers pwn your webapp. Parsing HTML with regex summons tainted souls into the realm of the living. HTML and regex go together like love, marriage, and ritual infanticide. The center cannot hold it is too late. The force of regex and HTML together in the same conceptual space will destroy your mind like so much watery putty. If you parse HTML with regex you are giving in to Them and their blasphemous ways which doom us all to inhuman toil for the One whose Name cannot be expressed in the Basic Multilingual Plane, he comes.

You should really click through to see the Stack Overflow post since there’s some additional visual humor from the markup and unicode characters used. Programmer Jeff Atwood has a companion post comparing parsing HTML with regular expressions to succumbing to the temptations of the dark god Cthulhu.

The point of both posts is quite simple: not only is parsing HTML with regular expressions incredibly difficult, most of the time you’re solving a problem that’s already been solved by hundreds of HTML/XML parsers.

HTML parsers have the advantage of knowing about the pitfalls of real HTML as it exists in the real world. A good parser will allow for malformed HTML and can stumble its way through some pretty ugly code and still pull out the data you want or perform whatever changes you’d like to make.

Of course not every bit of HTML will require a full parsing engine. The important thing to remember when deciding which tool to use is that there is a big difference between parsing and simply extracting HTML. If you just want to extract a line from a static webpage, a regex pattern might be the best tool.

On the other hand if you need to sanitize HTML, parse it and then traverse and manipulate it you will go mad using regular expressions. And that’s just silly when using a good HTML parser can eliminate the maddening hassles of maintaining complex regular expressions to handle every weird HTML situation you encounter.

When it comes to HTML parsers, Perl aficionados have a wide range of options. Python fans can use Beautiful Soup, Ruby has Nokogiri, among others. PHP users can try PHP Simple HTML DOM Parser, QueryPath and others. Microsoft’s .NET offers the HTML Agility Pack and C users can try Module HTMLparser [Update: Java fans can check out: HTMLParser]. Even JavaScript can handle the task pretty well with John Resig’s HTMLParser. In short, there’s a good HTML Parser out there in the language of your choice; you don’t need to reinvent the wheel.

We’re not trying to say you should never parse HTML with regular expressions. There are no absolutes in real world programming, but don’t reach for the regex first. Chances are, your problems are easier to handle with a parser.

See Also:

File Under: APIs, Location, Social, Web Services

Twitter Puts Geotagging Tools in Place

Twitter has announced a new feature that will give users the ability to send their location with each tweet.

The new geotagging API tools are available only through the Twitter API, and are not part of Twitter.com just yet. So, if you’d like to post geodata with your tweets, you’ll need to use a third-party application that supports the new features. Expect updates to the most forward-looking clients soon.

The geodata tools are also opt-in, meaning that you’ll have to head to your account settings and check “enable geotagging” before any application is allowed to send geodata with your tweets. If broadcasting your location strikes you as an invitation to an Orwellian nightmare, just ignore the new settings.

Judging by the way Twitter has set up the API, the geotagging tools are intended mainly for mobile applications running on platforms like the iPhone or Android, both of which have GPS or similar sensors for detecting location and can offer geodata to applications. Approve the Twitter app of your choice to access your location and it can then send the geotags along to Twitter.

In addition to the obvious — the ability to search for nearby tweets — the added geo information opens up a whole new realm of Twitter mashups — travel guides, local music searches and most likely quite a few things no one has thought of yet.

It also serves as both a boon and a challenge to the various location-based games and services that have sprung up along Twitter’s coral reef. Services like Foursquare, Loopt and Gowalla, which exist as separate apps, but incorporate Twitter for passing status messages, will have the ability to geotag those status messages. Likewise, you could run a simple Twitter search for messages geotagged with the location of the restaurant you’re currently sitting in and get recommendations from other Twitter users about which menu items to try.

Of course given the opt-in nature of geotags (a good decision on Twitter’s part), searches looking at only geotagged tweets will likely be a minority sampling of what’s actually happening on Twitter.

Developers interested in using the new API in applications and mashups should check out the new documentation.

See Also:

Microsoft Still Chasing the Competition With IE9

Serious work has begun on Internet Explorer 9, the next revision of Microsoft’s flagship web browser.

That sounds like good news, right? After all, IE8 has its moments, but it isn’t exactly a cutting-edge browser. Certainly, any improvement would seem welcome.

Yet, judging by the reaction from the web-development community on Microsoft’s IEBlog, you’d think Microsoft just announced the release of a major virus.

To understand why web developers — and even ordinary users — aren’t particularly thrilled with this early preview of IE9, we need to start by taking a look at IE8′s shortcomings:

  • Speed — This is all that matters for the average user, and all of IE8′s competitors are faster, something even Microsoft doesn’t deny.
  • Emerging standards — Firefox, Safari, Chrome and Opera have all begun implementing support for HTML5 and CSS 3, while IE8 has not. As more and more web apps take advantage of HTML5 tools, IE is in danger of becoming a second-class citizen on the web.
  • Web apps — In addition to lagging in overall page-rendering speed, IE8 is well behind the competition when it comes to JavaScript performance. Though Microsoft has been quick to challenge the relevance of JavaScript benchmarks, regular users of Gmail, Facebook and other JavaScript-heavy web apps do not.

Now let’s take a look at what improvements Microsoft is planning to make in IE9.

Speed

The first item of business on the IEBlog post is IE9′s speed improvements. There are two basic elements, page-rendering times (including JavaScript improvements) and a proposed hardware-acceleration layer that hands off complex rendering tasks to the graphics card.

After a rather lengthy treatise on why JavaScript benchmarks aren’t really an accurate measure of page-load speed, Microsoft goes on to tout IE9′s improved JavaScript performance. Microsoft offers a graph of IE9 running the SunSpider JavaScript test, a common way of measuring JavaScript performance.

The results are split over two graphs, one with IE8 versus the browsers its competitors are currently shipping, and the other charting IE9 against other experimental builds.

However, what’s really interesting is combining the two graphs. Doing so shows IE9′s JavaScript speed is roughly on par with Firefox 3.5, but still much slower than Safari 4 and Chrome 3.

Microsoft’s chart showing JavaScript rendering speeds in various browsers. Shorter bars are better.

Why advertise the fact the latest and greatest builds of Internet Explorer still can’t beat the actual shipping versions of the competition? Frankly, we’re not sure. But we assume Microsoft plans to continue improving IE9 before it finally ships. Unfortunately for IE9, we assume Mozilla, Apple and Google plan to do the same with their experimental builds.

And that cuts to heart of why developers and anyone with an interest in the using the web of the future today has long since lost faith in Internet Explorer: The competition continues to deliver improvements at a pace that far outstrips Internet Explorer.

Standards and HTML5

While speed is probably the most obvious and important feature of a web browser, the faster development time of IE’s competitors also means they are able to add new, experimental features long before IE.

That’s why Firefox, Safari, Opera and Chrome already have support for large portions of HTML5 and CSS 3, while IE 8 has next to none.

IE8 saw Microsoft catching up and finally getting the basics of HTML 4.x and CSS 2.1 right (we’ll overlook IE8′s lack of support for CSS pseudo element syntax), but unfortunately for IE8, the web is already moving on to HTML5 and CSS 3.

The good news is that IE9 will finally support most of CSS 3. There’s a screenshot on the IEBlog that appears to show IE9 rendering 41 out of 43 selectors in the CSS 3 selector test.

That’s great news for web developers, because it means less work building standards-based websites — provided IE9 delivers on this front.

However, when it comes to HTML5 support, IE9 appears decidedly less progressive. Microsoft appears to be sticking to its rather hard line on HTML5 — it’s not an official recommendation, so we’re not going to build support for it until it is.

While Microsoft is technically right about HTML5 (it is expected to become a recommendation in about a year), the truth is the web moves at the speed of the people actually building and using it, not the speed of recommendations from the W3C. At this rate, the lack of HTML5 support is looking more and more like Internet Explorer’s death knell.

The IEBlog does mention the HTML5 storage API, which was included in IE8, but ignores other elements already enjoying support in IE’s competition. For example, there’s no mention of HTML5′s audio, video or canvas tags, nor is there any discussion of the Geolocation API, Web Workers or SVG tools.

The thing to remember is that HTML5 support isn’t just a question of making web developers happy. If Microsoft wants IE to continue to be relevant to the future of the web, it’s going to have to step up its HTML5 support. The lack of support for the emerging standard gives Google a great way to attack IE — simply build sites that don’t work in IE and offer a link to download Chrome Frame.

That’s exactly what happens if you try logging into Google Wave with IE8. Clearly, Google and others are planning to use HTML5 with or without IE at the party. The short story, from what Microsoft has revealed thus far, is that IE9′s standards support will be catching up to where Firefox, Safari and Opera were two or three years ago.

Other Features

The IEBlog also touts the fact that IE9 will use Windows’ DirectX APIs to move graphics and text rendering from the CPU to the graphics card using Direct2D and DirectWrite. That means that IE 9 should be faster at rendering pages, particularly on PCs that have more-powerful graphics cards.

Of course, once again, the competition is already moving in the same direction. In most cases, the other browsers are using WebGL, which handles not just 2-D rendering, but also 3-D as well.

The IEBlog also touts IE9′s improved text-handling with sub-pixel positioning and much better anti-aliasing. Again, nice to see IE9 catching up with the competition.

Conclusion

Microsoft needs to hit a home run with IE9, or the IE franchise is going to go the way of Geocities. Unfortunately, based on what Microsoft has shown so far, IE9 looks to be a base hit at best. Certainly IE 9 will be good news on several fronts, notably the speed improvements and the increased CSS 3 support. But once again IE is catching up, not leading the way as it once did.

The typical rebuttal to IE’s shortcomings is that it doesn’t matter — IE still maintains a dominant market share, and will continue to do so, because it ships alongside Windows on new computers. It’s true that IE controls a majority share of the web. Microsoft got that majority because it bested the competition. Keep in mind that IE’s majority share used to be much, much larger, and it continues to slip with every passing month.

While we’re sure there are plenty of people who would love to dance on IE’s grave, the truth is that competition is a good thing. We want to see Microsoft make a better browser. Sadly, thus far, IE9 doesn’t look very competitive.

See Also:

File Under: Browsers, Software, Web Apps

Video: Google Offers Overview of Chrome OS

“What if your browser was your operating system?”

That’s the question Google is hoping to answer with Chrome OS, the open source operating system centered entirely around its Chrome browser. All your apps — e-mail, communications, docs, photo management — are delivered through the web browser, which sits on a lightweight Linux-powered desktop.

The company debuted Chrome OS Thursday morning during a press event at its headquarters in Mountain View, California. Wired.com’s Dylan Tweney was there to bring us all the details, and you can expect a post from him Thursday morning [Update: Dylan's post is up on Gadget Lab].

In the meantime, Google has posted a video (below) making an argument for why you’d want an operating system that funnels all of your productivity tasks through a browser: “It just gives you the internet, which is all most of us use our computers for now, anyways.”

And here’s a demonstration of the user interface. It’s striking — the browser is all there is to it.

Read the overview on the Google Blog. Also, be sure to check out some of the initial design specs, including the OS’s ability to auto-update silently and daily.

See Also:

File Under: Browsers, Security, Software

Firefox 3.6 Beta 3 Gains Security Features, Loses Windows 7 Integration

Mozilla has released a third beta for Firefox 3.6 with more than 90 bugfixes since beta 2, which was released just last week. If you’d like to take beta 3 for a spin, head over to the Mozilla downloads page.

Although beta 3 doesn’t contain any significant new features, it does have some welcome bug fixes and is considerably more stable than the previous betas. There is one feature not found in previous releases — add-ons can now access Firefox’s built-in geo-location features.

Unfortunately for Windows 7 users, much of the Windows 7 integration — like Aero tab previews and jump lists — has been removed. It remains to be seen whether or not those features will make it in the final release or will be postponed for Firefox 3.7.

The good news is that more than half of all add-ons now work with Firefox 3.6, including the recently released Weave update and other popular add-ons like Ad Block Plus and Firebug.

One big change on Firefox’s backend being introduced in beta 3 is a new restriction on how third-party add-ons integrate with Firefox. The Firefox components directory is now off limits to third-party tools. According to the Mozilla Developer Blog, “there are no special abilities that come from [accessing the components directory].”

The move is mainly designed to make Firefox more stable by preventing add-ons from accessing lower level tools that could cause crashes.

As the Mozilla Links blog points out, current Firefox 3.6 nightly builds are labeled as “preb4,” which might mean we’ll see a fourth beta before Firefox 3.6 arrives in final form. If Mozilla continues to crank out new betas every week, look for beta 4 around Thanksgiving with the final release arriving during December.

See Also: