Group items tagged

Waiting for GPhone to ring? MATHEW INGRAM Globe and Mail Update E-mail Mathew Ingram | Read Bio | Latest Columns September 6, 2007 at 12:01 AM EDT Apple may be the undisputed king of preannouncement hype (with the recent flurry of rumours about new iPod features, some of which were announced yesterday, being just the latest example) but Google has to run a close second. And the current hot spot for die-hard fans of the technology giant is the much-rumoured “Google phone” or GPhone. Whispers about such an animal have been circulating in the geekosphere since last year, but have recently caught fire again, with reports that some Silicon Valley insiders have seen actual prototypes of such a phone, and other reports that Google might be ready to launch a version in India. Om Malik, a former writer for Business 2.0 magazine who runs a blog network called GigaOm, recently spoke to a technology insider who said the tech company is working on a Google Phone that will run a modified version of the Linux operating system and have a special Web browser. According to some analysts, Google has been talking with several major carriers about distributing a low-price (or even free) device, the cost of which would be paid for by advertising. Various reports – including one in The Wall Street Journal last month – have said the phone will be loaded with Google software, including Gmail, Google Talk, an instant messaging and voice-calling service, and a special version of its mapping software that would use the device's built-in GPS. Skeptics (including yours truly) have pooh-poohed the idea that Google might want to get into the phone business, in the sense of designing and/or selling a specific piece of hardware. Why? Because it seems antithetical to what Google has been about – namely, perfecting software for searching, and more recently for search-related ads, e-mail and other online services. The only piece of hardware that Google makes or is involved in making is the Google “appliance” that companies can use as part of their internal computer networks – although it also plays a role in customizing the hundreds of thousands of servers in its giant server “farms.” It wasn't that long ago that Google was rumoured to be building a cheap “Google PC,” one that would be loaded with Google software and sold through retailers such as Wal-Mart. That never came to pass, however, and the company made a point of saying it had no intention of getting into the notoriously cutthroat hardware business. One of the driving forces behind much of the Google Phone speculation appears to be dissatisfaction with existing cellphone companies and carriers. You can almost hear people wishing that a giant company with billions of dollars in cash – not to mention a propensity for offering things for free – would come in and shake up the mobile phone market. The GPhone rumours have been persistent in part because there are so many puzzle pieces that appear to fit, with one of the biggest being a Google employee named Andy Rubin. A co-founder of Danger Inc., the company that developed the Sidekick phone/PDA, Rubin started another company called Android, which was acquired by Google in 2005. Google also bought a Waterloo, Ont.-based mobile software company called Reqwireless. Google recently said that if the U.S. Federal Communications Commission were to change the way it auctions wireless spectrum, the company would be willing to spend almost $5-billion (U.S.). And there have been repeated rumours about Google buying up “dark” or unused fibre capacity across the U.S. The perfect foundation for a new Google Phone service, some say. But does this mean a dedicated piece of hardware known as the Google Phone is definitely coming? Hardly. There's no question that Google has been actively courting phone makers and carriers about bundling its applications on their devices. It may even be working on developing its own mobile phone software, to make those Web applications easier to develop and use. But it's still quite a leap from that to a Google Phone. Of course, when you are churning out billions of dollars in free cash flow every month the way Google is, virtually anything is possible. Mobile phone users who are craving something new – and can't afford an iPhone – will no doubt be keeping their fingers crossed.

Paul, Jake and I were chatting a few weeks ago wondering how we can establish an ongoing dialog with our peers in product strategy and capture the innovative ideas they have for our future products. We thought of several ways to do this: Having conference calls to exchange ideas on a regular basis Inviting our peers to collaborate on a Google doc Build a simple website to track their ideas Obviously, 1 and 2 are bad ideas, so we opted for 3. What we needed was a site where people can submit their ideas, tag them, have them be rated by their peers, and allow comments to be entered. I like to call it “The Wall” — throw your ideas up on the wall and see if they stick… the community will decide. The real name of the site is “IdeaFactory”. A pretty simple site, really. It follows the same principles as most Web 2.0 sites today — folksonomies and feedback systems (ratings/comments) to facilitate community building. So, that night I set out to build it. As you know, I’m a rails fanboy now and so building it with RoR was a no brainer for me. I already knew that there were rails plugins for most of the features we needed (tagging, comments, search, etc.) My job would be to just tie them all together. To start, I gathered together the best plugins I knew about to build this site (I used agilewebdevelopment.com to help me). I ended up with these plugins: acts_as_commentable — for comments integration acts_as_ldap_authenticated — this is a variation on acts_as_authenticated with LDAP authentication support. In the future, I think I’ll migrate the LDAP code in this plugin to the restful_authentication plugin. I needed this plugin to tie into Oracle’s LDAP system so that users can just use their Oracle userid/pwds to get into the site. acts_as_taggable_on_steroids — for tagging support asset_packager — not necessary, but does a nice job of combining and minifying my javascripts and stylesheets minus_r — not necessary, but I hate the way rails treats javascript (they make you code your javascript in ruby… lame). Also, I wanted this since I prefer to use jQuery instead of Prototype. permalink_fu — not necessary, but gives me nice readable URLs acts_as_rateable — enables a five star rating system tiny_mce — enables WYSIWYG text editing which allows people to enter their content with some basic formatting. The beauty of using rails is that over the past few years, it’s become a popular choice for building “2.0″ style apps. And so, lots of the features of a “2.0″ style web application have been turned into rails plugins which makes building stuff with those features dirt simple. It’s also a framework that has a huge (and growing) community of developers who love to share their knowledge and code. When I started building the IdeaFactory, I had no idea that I would have a working version within 24 hours with all the key feature (tagging, ratings, comments, and LDAP auth). I’ve built a few rails apps before this one, but none that were really that interesting. The IdeaFactory is something that was interesting because it was badly needed by our teams — too many ideas weren’t being shared and critiqued by the general Oracle ecosystem. So, we knew that if we built the IdeaFactory, it would get used a fair bit and would help Oracle product strategists be more collaborative. I started coding on a Thursday night and by mid-day Friday morning, I had the general pieces in place so that data can be entered. On Friday afternoon, I requested a new hostname (http://ideas.us.oracle.com — intranet) which came alive by Saturday. I made a few enhancements over Saturday and Sunday and by Monday, the site was live! Since then, the site’s taken off (thanks to the additional boost by Justin). It’s become such a popular site internally, that there’s talk of putting together a public facing IdeaFactory site for Oracle customers — I’m hoping that happens. While many of us in development have been used to the whole process of requirements gathering, writing a BRD (business requirements doc), FDD (functional design doc), and TDD (technical design doc) — which I’m intimately used to doing over the years, it’s refreshing to be able to just roll up my sleeves and start building something and have a working product within hours of starting. I can’t wait for my next project. Stay tuned… you’ll hear about it here.

Avvo's new competitor It has been a few weeks since I mentioned Avvo, the controversial Seattle online attorney rating service that was promptly sued after its debut in June. Now, it turns out that Avvo will be facing some competition from an Arlington, Va.-based startup called HireTrade that has created a "legal marketplace" in which attorneys can post online profiles and clients can attempt to find attorneys to work on their cases. For now, the company is focusing on the legal markets in Washington, D.C., New York and Philadelphia, but it plans to expand to other geographic areas in the future. HireTrade says its HourlyValue Ratings System taps the knowledge of clients to rate attorneys, a method that it believes is more effective than a 1 to 10 rating scale like what Avvo employs. (HireTrade says it will not post a rating of an attorney until it accumulates three projects/reviews of the attorney.) At the present time, HireTrade's database does not appear to be as robust as Avvo's. I searched for three different attorney names and did not get a result or an attorney rating. (Similar searches on Avvo returned hundreds of attorneys.) Still, I asked HireTrade Chief Executive Neil Sandhu how his company is different from Avvo. Here's what he said: We think that existing professional ratings systems, including most numerical ratings systems, result in arbitrary values as it is impossible for a numerical algorithm to really capture every single nuance of a professional's background or decide what is the correct weight to place on different items of information such as experience, education, etc. (not to mention the fact that it appears that some of these systems merely assign higher scores to those professionals who take the time to fill out their profiles in more detail). Thus, meaningful comparisons are difficult to draw from the numerical values that result. Furthermore, there is very little actual intrinsic value in any number generated by such a system. And he continued: But what does a 5.0, 7.5, or 10 really signify when it comes to the seller of a service and, even if a numerical system was somehow perfect at capturing and interpreting data, can you really effectively compare someone who has a 8.8 to someone who has a 9.5--what if the person with an 8.8 has worked on more difficult matters and the person with the 9.5 has worked a long time and done very well with less difficult work? When the service is at the heart of what is being sold, we think an entirely different system must be used. Those are also some of the arguments made in the lawsuit against Avvo. Posted by document.writeln(showE2("johncook","seattlepi.com","John Cook"))John Cook John Cook at August 27, 2007 11:35 a.m.Category: Avvo Comments #48664Posted by unregistered user at 8/27/07 8:33 p.m.hey. i told those guys to turn their site into a myspace for lawyers but nooooo. what do i know Report violation #48671Posted by unregistered user at 8/27/07 9:31 p.m.Avvo's primary competition is certainly their own misguided idea and the shark who pounced on their inane offering. Report violation

Monday, October 29, 2007 The Semantic Web Goes Mainstream Continued from page 1 By Kate Greene Print E-mail Audio » New! Listen - Flash Listen - MP3 Subscribe to podcast What is this? Powered by Share » Digg this Add to del.icio.us Add to Reddit Add to Facebook