Regular readers of this blog may have noticed that accessing this blog has been flaky all weekend. I've gotten numerous reports via Twitter that my blog was displaying the following error message when being visited

Server Error in '/site1/weblog' Application.
--------------------------------------------------------------------------------
Exception of type 'System.OutOfMemoryException' was thrown.
Description: An unhandled exception occurred during the execution of the current web request. Please review the stack trace for more information about the error and where it originated in the code.
Exception Details: System.OutOfMemoryException: Exception of type 'System.OutOfMemoryException' was thrown.
Source Error:
An unhandled exception was generated during the execution of the current web request. Information regarding the origin and location of the exception can be identified using the exception stack trace below.
Stack Trace:
[OutOfMemoryException: Exception of type 'System.OutOfMemoryException' was thrown.]
Go1999(RegexRunner ) +0
System.Text.RegularExpressions.CompiledRegexRunner.Go() +14
System.Text.RegularExpressions.RegexRunner.Scan(Regex regex, String text, Int32 textbeg, Int32 textend, Int32 textstart, Int32 prevlen, Boolean quick) +144
System.Text.RegularExpressions.Regex.Run(Boolean quick, Int32 prevlen, String input, Int32 beginning, Int32 length, Int32 startat) +134
System.Text.RegularExpressions.Regex.Match(String input) +44
newtelligence.DasBlog.Web.Core.UrlMapperModule.HandleBeginRequest(Object sender, EventArgs evargs) +458
System.Web.SyncEventExecutionStep.System.Web.HttpApplication.IExecutionStep.Execute() +68
System.Web.HttpApplication.ExecuteStep(IExecutionStep step, Boolean& completedSynchronously) +75

This seems like a straightforward issue. The machine my blog is hosted on is running out of memory probably because the server is either overloaded from having too many sites hosted on it or one of the sites is badly written and is using up more than its fair share of memory. This has happened once or twice in the past and when it did I used the Live Chat feature of WebHost4Life, my hosting company, to contact a support rep who quickly moved my site to a different server. Thus when I got the first reports of this issue I thought this would be a routine support issue. I was sorely mistaken.

Earlier this year, WebHost4Life started migrating their customers to a new infrastructure with a different set of support staff. It seems this was a cost cutting measure because the new support staff seem to be a lot less technical than their previous counterparts and seem to have less access to their infrastructure. Over the weekend I chatted with about three or four different support folks over IM and opened a ticket that was closed multiple times. Here is a sampling of some of the messages that were written as part of closing the support ticket by WebHost4Life

Hello,

Thank you for contacting Support.

We apologize for any inconvenience this has caused you. I have checked your website at the URL: http://www.25hoursaday.com/weblog/ and it is working fine. Could you please check it once again after clearing browser cache and cookies? If the issue persists, please get back to us with the exact error message, so that we will investigate on your issue.

Thank you!

Sincerely,

Sharon <redacted>
Customer Support

~~~ TICKET CLOSED

Hello,

Thank you for your reply.

I have checked your website URL: http://www.25hoursaday.com/weblog/ and I was able to duplicate the issue. I have received the error message. Hence, I have asked a member of our team who specializes in website management to review your issue. You should be hearing from this specialist within 24-48 hours. If you have any questions in the meantime, please let us know and be sure to refer to the link http://www.webhost4life.com/member/sconsole for the quickest service.

Thank you for choosing webhost4life, we appreciate your support.

Sincerely,

Sharon <redacted>
Customer Support

~~~ TICKET ESCALATED TO TIER 2 SUPPORT

Hello,

I have checked the issue and and was not able to duplicate it. It seems to be issue with your ISP. Please check once again and if the issue still persists, please get back to us, tracert results of the website when you experience the issue and also the exact time and location so that we can investigate on it further.

If you have any further questions, please update the Support Console.

Sincerely,

Aleta <redacted>
Technical Specialist

~~~ TICKET CLOSED

Hello,

Thank you for getting back to us.

Currently, your website http://www.25hoursaday.com/weblog/ is loading without any slowness. I suggest you to check the website functionality again and get back to us, if the issue persists.

If you have any further questions, please update the Support Console.

Sincerely,

Kara
Technical Specialist

~~~ TICKET CLOSED

I’m still getting reports that my blog is throwing out of memory exceptions infrequently and as you can see from the above there isn’t anything being done to fix this problem. At first I wondered if the poor support I was getting was because this happened on the weekend and perhaps the technical folks only work weekdays. Unfortunately my hopes were dashed when someone on Twitter pointed me to webhosting review site with dozens of complaints about WebHost4Life which echoed my experience. It seems the company is under new management and quality has suffered under some of their cost cutting moves.

Although I paid for a year’s worth of service, I’ve decided that I probably need to switch hosting companies. Thus I’m seeking recommendations for a web hosting company that supports ASP.NET and .NET 2.0 or higher from any of my readers. Also if anyone is familiar with the process of cancelling your service with a web host including getting refunded for services not provided, I’d love any experience or tips you have to share.

I've been a Twitter user for almost two years now and I have always been impressed by the emergent behavior that has developed from simply giving people a text box with 140 character limit. The folks at Twitter have also done a good job of noticing some these emergent behaviors and making them formal features of the site. Both hashtags and @replies are examples of emergent community conventions in authoring tweets that are now formal features of the site.

Twitter recently added retweets to this list with Project Retweet. After using this feature for a few days I've found that unlike hashtags and @replies, the way this feature has been integrated into the Twitter experience is deeply flawed. Before I talking about the problems with Project Retweet, I should talk about how the community uses retweeting today.

Retweeting 101: What is it and why do people do it?

Retweeting is akin to the practice of forwarding along interesting blog posts and links to your friends via email. A retweet repeats the content of a person's tweet (sometimes edited for brevity) along with a reference to the user who is being retweeted. Often times people also add some commentary to the retweets. Examples of both styles of retweets are shown below.

Figure 1: Retweet without commentary

Figure 2: Retweet with added comment

Unlike hashtags and @replies, the community conventions aren't as consistent with retweets. Below are two examples of retweets from my home page which use different prefixes and separators from the one above to indicate the item is a retweet and the user's comment respectively.

Figure 3: Different conventions in retweeting

However there are many issues with retweeting not being a formal feature of Twitter. For one, it is often hard for new users to figure out what's going on when they see people posting updates prefixed with strange symbols and abbreviations. Another problem is that users who want to post a retweet now have to deal with the fact that the original tweet may have taken up all or most of the 140 character limit so there may be little room to credit the author let alone add commentary.

Thus I was looking forward to retweeting becoming a formal feature of Twitter so that these problems would be addressed. Unfortunately, while one of these problems was fixed more problems were introduced.

Flaw #1: Need to visit multiple places to see all retweets of your content

Before the introduction of the retweet feature, users could go to http://www.twitter.com/replies to see all posts that reference their name which would include @replies and retweets. The new Twitter features fragments this in an inconsistent manner.

Figure 4: Current Twitter sidebar

Now users have to visit http://www.twitter.com/replies to see people who has retweeted their posts using community conventions (i.e. copy and pasting then prefixing "RT" to a tweet) and then visit http://twitter.com/retweeted_of_mine to see who has retweeted their posts by clicking the Retweet link in the Twitter web user interface. There will be different people in both lists.

Figure 5: Retweets in the Replies/Mentions page

Figure 6: Retweets on the "Your tweets, retweeted" page

It is surprising to me that Twitter didn't at least include posts that start with RT followed by your username in http://twitter.com/retweeted_of_mine as well.

Flaw #2: No way to add commentary on what you are retweeting

As I mentioned earlier, it is fairly common for people to retweet a status update and then add their own commentary. The retweet feature built into Twitter ignores this common usage pattern and provides no option to add your own commentary.

Figure 7: The Retweet prompt

This omission is particularly problematic if you disagree with what you are sharing and want to clarify to your followers that although you find the tweet interesting you aren't endorsing the opinion. 

Flaw #3: Retweets don't show up in Twitter apps

One of the other surprising changes is that Twitter retweets have been introduced in a backwards-incompatible manner into the API. This means that retweets created using the Twitter retweet button do not show up in 3rd party applications that use the Twitter API. See below for an example of what I see in Echofon versus the Twitter web experience and notice the missing tweet.

Figure 8: Twitter website showing a retweet

Figure 9: The retweet is missing in Echofon

Again, I find this surprising since it would have been straightforward to keep retweets in the API and exposing them as if they were regular old school retweets prefixed with "RT".

Flaw #4: Pictures of people I don't know in my stream

The last major problem with the Twitter retweet feature is that it breaks user expectation of the stream. Until this feature shipped, users could rest assured that the only content they saw in their stream was content they had explicitly asked for by subscribing to a user. Thus when you see someone in your stream the person's user name and avatar are familiar to you.

With the new retweet feature, the Twitter team has decided to highlight the person being retweeted and treat the person who I've subscribed to that did the retweeting as an afterthought. Not only does this confuse users at first (who is this person showing up in my feed and why?) but it also assumes that the content being retweeted is more important than who did the retweeting. This is an unfortunate assumption since in many cases the person who did the retweeting adds all the context.

Now Playing: Jason Derulo - Whatcha Say

I've been a FeedBurner customer for a couple of years and was initially happy for the company when it was acquired by Google. This soon turned to frustration when I realized that Google had become the company where startups go to die. Since being acquired by Google almost two years ago, the service hasn't added new features or fixed glaring bugs. If anything, the service has only lost features since it was acquired.

I discovered the most significant feature loss this weekend when I was prompted to migrate my FeedBurner account to using a Google login. I thought this would just be a simple change in login credentials but I got a different version of the service as well. The version of the service for Google accounts does have a new feature, the chart of subscribers is now a chart of subscribers AND "reach". On the other hand, the website analytics features seemed to be completely missing. So I searched on the Internet and found the following FAQ on the FeedBurner to Google Accounts migration

Are there any features that are not available at feedburner.google.com?

There are two features that we are retiring from all versions of feedburner: Site Stats (visitors) and FeedBurner Networks.

We have decided to retire FeedBurner website visitor tracking, as we feel Google already has a comparable publisher site analytics tool in Google Analytics. If you do not currently use Google Analytics, we recommend signing up for an account . We want to stress that all feed analytics will remain the same or be improved, and they are not going away.

FeedBurner Networks, which were heavily integrated with FeedBurner Ad Network, are no longer being supported. As with many software features, the usage wasn't at the level we'd hoped, and therefore we are making the decision not to develop it further, but to focus our attention on other feed services that are being used with more frequency. We will continue to look out for more opportunities for publishers to group inventory as part of the AdSense platform.

I guess I should have seen this coming. I know enough about competing projects and acquisitions to realize that there was no way Google would continue to invest in two competing website analytics products. It is unfortunate because there were some nice features in FeedBurner's Site Stats such as tracking of clicked links and the ability to have requests from a particular machine be ignored which are missing in Google Analytics. I also miss the simplicity of FeedBurner's product. Google Analytic is a more complex product with lots of bells and whistles, yet it takes two or three times as many clicks to answer straightforward questions like what referrer pages are sending people to our site. 

The bigger concern for me is that both Google Analytics and FeedBurner (aka AdSense for Feeds) are really geared around providing a service for people who use Google's ad system. I keep wondering how much longer Google will be willing to let me mooch off of them by offloading my RSS feed bandwidth costs to them while not serving Google ads in my feed.  At this rate, I wouldn't be surprised if Google turns FeedBurner into a Freemium business model product where users like me are encouraged to run ads if we want any new features and better service.

Too bad there isn't any competition in this space.

Now Playing: Lords of the Underground - Chief Rocka

Every once in a while I encounter an online service or Web site that is so irritating that it seems like the people behind the service are just in it to frustrate Web users. And I don’t mean the obvious candidates like email spammers and purveyors of popup ads since they’ve been around for so long I’ve either learned how to ignore and avoid them.

There is a new generation of irritants and many of them are part of the new lunacy we call “Web 2.0”

1. Flash Widgets with Embedded PDF Documents: Somewhere along the line a bunch of startups decided that they needed to put a “Web 2.0” spin on the simple concept of hosting people’s office documents online. You see, lots of people would like to share documents in PDF or Microsoft Office® formats that aren’t particularly Web friendly. So how have sites like Scribd and Docstoc fixed this problem? By creating a Flash widgets containing the embedded PDF/Office documents like the one shown here. So not only are the documents still in a Web unfriendly format but now I can’t even download them and use the tools on my desktop to read them. It’s like let’s combine the FAIL of putting non-Web documents on the Web with the fail of a Web-unfriendly format like Flash. FAIL++. By the way, it’s pretty ironic that a Microsoft enterprise product gets this right where so many “Web 2.0” startups get it wrong.

2. Hovering Over Links Produces Flash Widgets as Pop Over Windows: The company that takes the cake for spreading this major irritant across the blogosphere is Snap Technologies and their Snap Shots™ product. There’s nothing quite as irritating as hovering over a link on your way to click another link and leaving a wake of pop over windows with previews of the Web pages at the end of said links. I seriously wonder if anyone finds this useful?

3. Facebook Advertisers: One of the promises of Facebook is that its users will see more relevant advertising because there is all this rich demographic data about the site’s users in their profiles. Somewhere along the line this information is either getting lost or being ignored by Facebook’s advertisers. Even though my profile says I’m married and out of my twenties I keep getting borderline sleazy ads whenever I login to play Scrabulous asking if I want to meet college girls. Then there are the ads which aren’t for dating sites but still use sleazy imagery anyway. It’s mad embarrassing whenever my wife looks over to see what I’m doing on my laptop to have dating site ads blaring in her face. Obviously she knows I’m not on a dating site but still…

4. Forums that Require Registration Showing Up in Search Results : Every once in a while I do a Web search for a programming problem and a couple of links to Experts Exchange end up in the results. What is truly annoying about this site is that the excerpt on the search result page makes  it seem as though the answer to your question is one click away but when you click through you are greeted with “All comments and solutions are available to Premium Service Members only”. I thought search engines had rules about banning sites with that sort of obnoxious behavior?

5. Newspaper Websites with Interstitial Ads and Registration Requirements: Newspapers such as the New York Times often act as if they don’t really want me reading the content on their Web site. If I click on a link to a story on the New York Times site such as this one, one of two things will happen; I’m either taken to a full page animated advertisement with an option to skip the ad in relatively small font or I get a one sentence summary of the story with a notice that I need to register on their Web site before I can read the story. Either way it’s a bunch of bull crap that prevents me from getting to the news.

There are two things that strike me about this list as notable. The first is that there are an increasing number of “Web 2.0” startups out there who are actively using Flash to cause more problems than they claim to be solving. The second is that requiring registration to view content is an amazingly stupid trend that is beyond dumb. It’s not like people need to register on your site to see ads so why reduce the size of your potential audience by including this road block? That’s just stupid.

Now Playing: Pleasure P - Rock Bottom (feat. Lil Wayne)