Last week there was an outage on NewsGator Online. This outage didn't just affect people who use the NewsGator Online but also users of their desktop readers such as FeedDemon which synchronize the users feed state between the desktop and the web-based reader.

In his post Dealing with Connectivity Issues in Desktop Applications Nick Bradbury writes

One of the more frustrating challenges when designing a desktop application that connects to the Internet is figuring out how to deal with connectivity issues caused by firewalls, proxy servers and server outages.
...
And as we discovered last week, when your application relies on a server-side API, it has to be able to deal with the server being unavailable without significantly impacting the customer. This was something FeedDemon 2.0 failed to do, and I have to take the blame for this. Because of my poor design, synchronized feeds couldn't be updated while our server was down, and to make matters worse, FeedDemon kept displaying a "synchronization service unavailable" message every time it tried to connect - so not only could you not get new content, but you were also bombarded with error messages you could do nothing about.

A couple of months ago I wrote a blog post entitled The Newsgator API Continues to Frustrate Me where I complained about the fact that Newsgator Online assumes that clients that synchronize with it actually just fetch all their data from Newsgator Online including feed content. This is a bad design decision because it means that they expected all desktop clients who synchronize with the web-based reader to have a single point of failure. As someone who's day job is working on the platforms that power a number of Windows Live services, I know from experience that service outages are a fact of life. In addition, I also know that you don't want clients making requests to your service unless they absolutely have to. This is not a big deal at first but once you get enough clients you start wanting them to do as much data retrieval and processing as they can without hitting your service. Having a desktop feed reader rely on a web service for fetching feeds instead of having it fetch feeds itself needlessly increases the costs of running your online service and doesn't buy your customers a significantly improved user experience. 

I've bumped into Greg Reinacker since I complained about the Newsgator API and he's been adamant about the correctness of their design decisions. I hope the fallout from the recent outage makes them rethink some of the design of Newsgator's RSS platform.

 

In the blog post entitled Rapleaf to Challenge eBay Feedback Mike Arrington talks about newly formed Rapleaf which aims to build a competitor to eBay's feedback system. This idea shows a lot of insight on the part of the founders. The value that eBay provides to sellers and buyers is primarily the reputation system and not as a venue for auctions. The network effects inherent in eBay's reputation system make it the ultimate kind of lock-in. No power seller or buyer will look at alternatives even if they are free (like Yahoo! Auctions) because they don't want to start from scratch with the reputation they've built or trust trading with people whose reputations haven't been built. However it isn't a slam dunk that Rapleaf will be a successful idea. 

In his post entitled Rapleaf's Fatal Flaws Ian McAllister of Windows Live Shopping writes

Flaw #1 - Transaction Unaware

Rapleaf is not in the middle of transactions. They have no way to determine if a transaction between two parties actually took place. Co-founder Auren Hoffman claims that their sophisticated human and machine-based fraud detection will be able to detect fraud but to me this seems like complete hand-waving...The success of eBay's feedback system rests completely on the fact that they attach feedback only to completed transactions where eBay collects money via commission.

 

Flaw #2 - Cold Start

Every new startup has a cold start problem and must build users, customers, partners, etc. from the ground up but Rapleaf has the mother of all cold start problems. The post mentioned nothing about how they plan to build mindshare in the market and I think they'll be dead in the water if they expect users to start going to www.rapleaf.com in droves all of a sudden and being keen to trust one of the 342 Rapleaf trusted sellers based on 2 items of feedback not attached to any verified transaction.

Flaw #2 was something I'd considered but Flaw #1 didn't even occur to me. Now that I consider it, I can't see how they can be successful as a competitor of companies like eBay since they aren't part of the transaction. It would seem to make more sense for them to be a partner of eBay except that there is no incentive for eBay to partner with them and thus provide an avenue out of the lock-in of eBay's feedback system.

Does this mean Rapleaf is DOA?

 

I read a number of news stories last week about Microsoft hiring a former exec from Ask.com to run MSN. A number of these news sources and corresponding blog posts got the story wrong in one way or the other

In her news story entitled Former Ask.com president will join Microsoft Kim Peterson of the Seattle Times wrote

Microsoft has hired the former president of search rival Ask.com to run its online business group, overseeing the MSN and Windows Live units and playing a big role in the company's move to the Web.

In the Reuters news story entitled Microsoft hires CEO of Ask.com to head Web unit it states

Software giant Microsoft Corp. said on Friday it hired away Steve Berkowitz, the chief executive of rival Internet company Ask.com, to head Microsoft's own Internet business.

In her blog post entitled CEO of Ask.com moves to Microsoft Charlene Li of Forrester Research wrote

Most importantly, Microsoft is taking a very important step in putting ALL of the hot consumer products under one team. Live.com is at the core of Microsoft's turnaround -- it represents fast development cycles and a totally new approach to addressing the marketplace. At the same time, Microsoft can't turn its back on the advertising juggernaut of MSN.com. In the past year, there's been uncertainty about how MSN.com and Live.com will work together. Having them all come together under Steve will be a first step in addressing the concerns of the MSN.com group while maintaining Live.com's momentum.

Highlighted in red are statements which are at best misleading. I'm not singling out the above news publications and bloggers, almost every article or blog post I read about Steve Berkowitz being hired gave the same misleading impression.

Why are they misleading? That's easy. Let's go back to the Microsoft press release Microsoft Realigns Platforms & Services Division for Greater Growth and Agility which breaks out Microsoft's internet business into the following three pieces

Windows and Windows Live Group
With Sinofsky in charge, the Windows and Windows Live Group will have engineering teams focused on delivering Windows and engineering teams focused on delivering the Windows Live experiences. Sinofsky will work closely with Microsoft CTO Ray Ozzie and Blake Irving to support Microsoft’s services strategy across the division and company.

Windows Live Platform Group
Blake Irving will lead the newly formed Windows Live Platform Group, which unites a number of MSN teams that have been building platform services and capabilities for Microsoft’s online offerings. This group provides the back-end infrastructure services, platform capabilities and global operational support for services being created in Windows Live, Office Live, and other Microsoft and third-party applications that use the Live platform. This includes the advertising and monetization platforms that support all Live service offerings.

Online Business Group
The new Online Business Group includes advertising sales, business development and marketing for Live Platforms, Windows Live and MSN — including MSN.com, MSNTV and MSN Internet Access. David Cole, senior vice president, will lead this group until his successor is named before his leave of absence at the end of April.

That's right, three pieces each with it's own corporate vice president. So Charlene Li isn't quite right when she says that MSN.com and Live.com are now aligned under Steve Berkowitz. Instead what's being aligned under him is the business development and marketing for both sites. The platform that powers Live.com should be under Blake Irving while the actual website development is under Steven Sinofsky.

I'm sure that makes as much sense to you as it does to me. However according to the press release, this organizational structure will increase Microsoft's agility in delivering innovation to customers.

I can't wait.

 

It's MS Poll season. This is when at work our employer encourages us to fill out an opinion poll on how we feel our day jobs and the company in general. Besides Mini-Microsoft I've seen a couple of the introspective posts about working in the B0rg cube I expect to see during this season such as Robert Scoble's How Microsoft can shut down Mini-Microsoft and Mike Torres's Playing to "not lose.

I don't really have anything intospective to add to what they've written. I probably won't fill out MS Poll this year since it's always felt to me like a pointless opinion poll. If my management can't tell what I like or dislike about working here then it's a screw up on both our parts which won't be fixed by a hastily filled out opinion poll.

I did find an interesting comment by Leah Pearlman to Mike Torres's post I felt compelled to talk about. She wrote

Re: Innovation. Hmm. I agree and disagree. I agree from the standpoint of a Microsoft employee who wants to work on innovative things.  I agree with you that there’s been too much talk about “how to beat the competition.” Reinventing the wheel because Yahoo! and Google have wheels doesn't get me out of bed in the morning. But! (you knew it was coming) My opinion lately has been that there's too much emphasis put on innovation at Microsoft, and it comes at the expense of fundamentals, intuitiveness, simplicity.  Often times there are great reasons why our competitors have done certain things , and I see people carelessly disregard these things in the name of innovation.

Innovation is one of those words Microsoft has killed. What has begun to irritate me is when people describe what is basically a new feature in their product as an innovation. To start off, by definition, since you work at a big software nothing you work on is innovative. Even Google who used to be raised up as the poster child of innovation in the software industry have been reduced to copying Yahoo! services and liberally sprinkling them with AJAX as they've grown bigger. 

Often when I hear people claiming that the new feature in their Microsoft product is an innovation it just makes them look ignorant. Most of their innovations are either (i) already shipping in products offered by competitors or startups that anyone who reads TechCrunch is aware of or (ii) are also being worked on by some poor slobs at AOL/Google/Yahoo! who also think their feature is extremely innovative. As Jeremy Zawodny pointed out in his post Secrets of Product Development and What Journalists Write

Larger companies rarely can respond that quickly to each other. It almost never happens. Sure, they may talk a good game, but it's just talk. Building things on the scale that Microsoft, Google, AOL, or Yahoo do is a complex process. It takes time.

Journalists like to paint this as a rapidly moving chess game in which we're all waiting for the next move so that we can quickly respond. But the truth is that most product development goes on in parallel. Usually there are people at several companies who all have the same idea, or at least very similar ones. The real race is to see who can build it faster and better than the others.

The culture of bragging about dubious innovations likely springs from the need to distinguish yourself from the pack in a reward culture that takes dog-eat-dog to another level. Either way, do me a favor. Stop calling your new features innovations. They aren't.

Thanks for listening

 

I've posted a number of blog  entries in the past about how popular various blogs on MSN Spaces are, especially the Asian ones. Unsurprisingly it's taken some of the folks from the insular geek blog set, a while to notice this trend. Recently, Scott Karp wrote about this in a blog post entitled Technorati Top 100 Is Changing Radically which was followed up by a blog post entitled Get on MSN Spaces in Asia and watch the link-love pile up. Sort of by Chris Edwards.

Both blog posts are interesting because the authors refuse to believe that it is possible for blogs they haven't heard of from Asian countries like China and Japan to be more popular than A-list technology bloggers like Dave Winer. In his post, Chris Edwards points out that the incoming links for blogs like M¥$ŤěяĬǾũ§ ĢÎѓĻ contain blogs that only link to the Space via the Recently Updated Spaces module. On an initial glance this seems to be true. When Technorati first started tracking MSN Spaces we realized this module would be a problem and added rel='nofollow' on all links to spaces from this module. This means that search engines and web crawlers should not consider these links as 'votes' for the site for page ranking purposes.

Ignoring that particular space, there are still a number of spaces in the Technorati Top 100 whose most recent links don't come from the Recently Updated Spaces module. For example, check out the incoming links to http://spaces.msn.com/MSN-SA, http://spaces.msn.com/atiger and http://spaces.msn.com/members/thespacecraft (MSN Spaces team blog).

As much as it seems to bother some technology geeks, a number of blogs hosted on MSN Spaces are more popular than so-called A-list technology bloggers.

 

From the Reuters article Microsoft heads to college to pitch Windows Live we get the following excerpt

The decision to outsource the University of Texas-Pan American's 17,000 student e-mail accounts to Microsoft Corp. for free was a simple one for Gary Wiggins, the school's top IT administrator.

Students hated the existing system and its limited storage, lack of features -- like a calendar, for example -- and cumbersome user interface.

"The legacy system we were moving from was so bad that the new features were very well-accepted," said Wiggins, who is the school's vice president for information technology.

The school could still create e-mail addresses ending in utpa.edu and many students were already familiar with Microsoft's Hotmail e-mail service.

The University of Texas Pan-American is not alone in linking up with Microsoft. The world's largest software maker has clinched deals to host e-mail systems for 72 institutions around the world and is in active discussions to add almost 200 more schools.

Microsoft sees its push onto college campuses as a way to promote its new Windows Live platform, an advertising-funded one-stop shop for Microsoft's Web services from e-mail to news to instant messaging to blogs.

The Windows Live @edu folks have done quite a bit over the past few months. I totally dig what we are doing with projects like theirs and Windows Live Custom Domains. I've actually started factoring in their scenarios when thinking about the next generation of Windows Live communication services we'll be building. The more Windows Live services we get to participate in this the better. Being able to give people email and IM accounts using the my own domain is a great first step but there are a bunch more things I'd like to see. 

 

One of the devs on Windows Live Favorites just snuck me the following screenshot

Sweet, huh?

 

Tim Bray has a post entitled The Cost of AJAX where he writes

James Governor relays a question that sounds important but I think is actively dangerous: do AJAX apps present more of a server-side load? The question is dangerous because it’s meaningless and unanswerable. Your typical Web page will, in the process of loading, call back to the server for a bunch of stylesheets and graphics and scripts and so on: for example, this ongoing page calls out to three different graphics, one stylesheet, and one JavaScript file. It also has one “AJAXy” XMLHttpRequest call. From the server’s point of view, those are all just requests to dereference one URI or another. In the case of ongoing, the AJAX request is for a static file less than 200 bytes in size (i.e. cheap). On the other hand, it could have been for something that required a complex outer join on two ten-million-row tables (i.e. very expensive). And one of the virtues of the Web Architecture is that it hides those differences, the “U” in URI stands for “Uniform”, it’s a Uniform interface to a resource on the Web that could be, well, anything. So saying “AJAX is expensive” (or that it’s cheap) is like saying “A mountain bike is slower than a battle tank” (or that it’s faster). The truth depends on what you’re doing with it. 

In general I agree with Tim that the answers to questions like "is technology X slower than technology Y" depends on context. The classic example is that it makes little sense arguing about the speed of the programming language if an application is data bound and has to read a lot of stuff off the disk. However when it comes to AJAX, I think that in general there is usually more load put on servers due to the user interface metaphors AJAX encourages. Specifically, AJAX enables developers to build user interfaces that allow the user to initiate multiple requests at once where only one requests could have been made in the traditional HTML model. For example, if you have a UI that makes an asynchronous request every time a user performs a click to drill down on some data (e.g. view comments on a blog post, tags on a link, etc) where it used to transition the user to another page then it is more likely that you will have an increased number of requests to the server in the AJAX version of your site. Some of the guys working on Windows Live have figured that out the hard way. :)

 

For some reason, the following story on Slashdot had me cracking up

Microsoft Plans Gdrive Competitor

Posted by samzenpus on Wednesday April 19, @09:13PM
from the personal-virtual-bill dept.
gambit3 writes "From Microsoft Watch: The MSN team is working on a new Windows Live service, code-named Live Drive, that will provide users with a virtual hard drive for storing hosted personal data. From early accounts, it sounds an awful lot like Gdrive, the still-as-yet-publicly-unannounced storage service from Google."

I have to agree with Mike Torres, 2006 is really 1998 in disguise. With the release of Google Page Creator, Google Finance, Google Calendar and the upcoming GDrive (aka Yahoo! GeoCities, Yahoo! Finance, Yahoo! Calendar and Yahoo! Briefcase knockoffs) it is now clear to me that Google's master plan is to become Yahoo! 2.0. On the other hand, with some of the recent Windows Live announcements we seem to be giving the impression that we are chasing Google's tail lights who in turn is chasing Yahoo! tail lights who in turn is chasing the tail lights of various 'Web 2.0' startups. Crazy.

I wonder what Google will do when they run out of Yahoo! services to copy?

 

Joe Gregorio has a post about the Google Data APIs Protocol where he points out that Google has released the Google Data APIs Protocol which for all intents and purposes, is an Atom Store (i.e. an Atom Publishing Protocol service mixed with OpenSearch). I took a glance at the Google Data APIs overview documentation and it states the following

GData is a new protocol based on Atom 1.0 and RSS 2.0.

To acquire information from a service that supports GData, you send an HTTP GET request; the service returns results as an Atom or RSS feed.You can update data (where supported by a particular GData service) by sending an HTTP PUT request, an approach based on the Atom Publishing Protocol.

All sorts of services can provide GData feeds, from public services like blog feeds or news syndication feeds to personalized data like email or calendar events or task-list items. The RSS and Atom models are extensible, so each feed provider can define its own extensions and semantics as desired. A feed provider can provide read-only feeds (such as a search-results feed) or read/write feeds (such as a calendar application).

On the surface, this looks like it aims to solve the same problems that Ray Ozzie's Microsoft's Simple Sharing Extensions for RSS and OPML aims to solve as well. At first glance, I think I prefer it to RSS-SSE because it is explicitly about two-way interaction as well as one-way interaction. RSS-SSE provides a good solution if I am a client application synchronizing information from a master source such as my online calendaring application but it is still unclear to me how I use the mechanics of RSS-SSE to push my own updates to the server. Of course, RSS-SSE is better for pure synchronization but GData looks like it would be a better fit for a generic read/write API for data stores on the Web, the same way RSS has become a generic read API for data stores on the Web.

This definitely has Adam Bosworth  written all over it. About a year ago, it was obvious from certain posts on the atom-protocol mailing lists that Adam had some folks at Google working on using the Atom Publishing Protocol as a generic API for Web stores. I'm surprised it's taken this long for the project to make its way out of the Googleplex. This would be a good time to familiarize yourself with Adam's paper, Learning from THE WEB.

PS: 'Google Data APIs Protocol' is a horrible name. You can tell that ex-Microsoft employees had a hand in this effort. ;)