Author: spierzchala

A couple of comments on the new MSN Spaces site.

1. Your web server headers for the main page are basically not helpful, especially the Caching ones:

   Cache-Control: no-cache
   Pragma: no-cache
   Expires: -1

   "-1" is not a valid Expires entry — jest set it to the current date. "Pragma" is a client-side header ONLY.

2. No compression? CSS, JS, and HTML file compression could save you some bandwidth and speed up the site.

The one site I looked at from an MSN staff member was an incredibly busy wild mess. But usability and site design are only things that I comment on.

Be interesting to see this try and evolve. Steve Rubel has a link to some of Steve Ballmer’s comments. He also links to a previous article that discusses the "Pearl Harbor" email in 1995 when BGates discovered the Web.

I think that this is a bit far-fetched on Mr. Rubel’s part. Unlike 1995, Microsoft does not carry the goodwill and universal support that followed it’s release of MSIE. Even if they get it right in version 3.0, Google, TypePad and other players will not allow them to dominate. They may be able to release a blogging product that integrates into IIS (oh wait! what database will it use…you need a license for SQL server too!), but that will only allow other companies to open in competition to the Spaces offering.

Yes, the MSN move may move blogging to a more mainstream audience, but the other players in the field will just move-in.

My thought is that 2005 will be the year of corporate consolidation in the blogosphere. Yahoo will acquire one of the major services (TypePad and Radio Userland come to mind), and Blogger will become more tightly integrated with the other Google offerings. It will be like the portal wars all over again.

Will be fun.

Tom Peters nails a comment on the comparison of food from the source versus distant interpretations.

I can’t wait for the delight that they call a Pancho Villa burrito next week — one of the benefits of going back to Silicon Valley next week.

Jakob Nielsen’s site is a great source for Web usability information. Too bad he doesn’t have a feed; I might visit his site more often.

His column — Alertbox — is an often scathing look at how Web sites use and mis-use design to drive their users mad.

Jakob, RSS isn’t that hard! Get a feed or be lost in the noise.

Interesting comments on how Microsoft is cranking up the FUD to handle Firefox.  Looks like Firefox is a credible threat to their core development platform.

For the first time since I left the Valley in October 2003, I will be returning there on business. It will be interesting to see how much it has/has not changed since we left.

I know that a lot of the people that I knew when I was there have been battered about by the storms of change, and that many of them have returned to where they came from. But, is Silicon Valley still as magic as the first time I arrived in 1999?

Can’t wait.

We want you as a new recruit…

Sometimes I wonder if the team that designed the CIA Web Site is managing the Marines Web site as well…

Stumbled across some research today outlining how the concept of Shortest Remaining Processing Time can be used to improve Web performance. (Google references here and here).

This is a very interesting concept that forces the Kernel (needs to be done on a system that supports TCP stack mods) to be unfair in its allocation of resources, by spitting out the requests that have the shortest time remaining first — in the case of a Web server, this means small files are sent before large files.

This is interesting, and not something that I have seen move beyond the academic literature. Is anyone out there using SRPT in their production Web serving environment? How did you determine that the predicted improvements in Web performance were in fact realized?

It also appears that their are some issues with SRPT, especially as file size increases. But it it would be interesting to get feedback from the Web performance community on this.

Today an e-mail came in from one of the sales reps in the field that just set me off. This person wanted us to generate a "best practices" document.

Under most circumstances, this would seem innocuous enough. But the whole idea that a simplistic, sales-oriented best practices document on Web performance could be produced that summarized these concepts in a few pages made me go ballistic.

I know what makes Web performance work. It took 5 years to get to this point, but after all of these years, I am burnt out trying to get the message across. The company I work for has a key vision. We have a strategy.

Tactically, there are far more Web sites with problems than there are that get it. I am not talking a small Web site trying to sell stuff on online auction sites or trying to reach a niche market. I am talking about large multinationals that are so busy in-fighting that they don’t see the dollars walking, marching, out the back door.

Business owners: poor Web performance costs you money. It is that simple.

But, I can’t get this message through to people anymore. All I want to do is help companies make their Web sites fast and reliable. And in the past year, I have been beating my head against the wall trying to do this.

I am passionate about this, like nothing else in my life. If your company feels the same way, I challenge you to re-ignite my passion.

The modern Web site is comprised of a number of complex applications designed to serve the needs of the business. Not only are they delivering content to end-users, but inventory information to partners via Web services, sales volume data to managers and executives, and revenue/cost data to the finance team.

All of this originates out of the Web site infrastructure, which is no longer a distinct vertical of the business, but a core component of the business. It’s no longer just about the housewife in Peoria, or Mumbai, or Berlin getting their item. The Web infrastructure is about the power to make key business decisions.

If the Web infrastructure is not looked at in this manner, then pack up your kit and go home. And how do you plan to ensure that it is running effectively?

The vast majority of IT departments while confidently tell a firm like the ones that I have worked for that they do not need external measurement and monitoring. Why? Because IT has never had to defend themselves to the business before. Now they are being asked to justify themselves before the business leaders. And in many cases, they don’t know how to.

It is not a battle. The business demands that their needs be met. IT needs the tools and technology to do this. External performance measurement is an ally in this justification. IT can clearly demonstrate what they are doing right, how they could do it better, or what they need to be more effective.

Business leaders need easy-to-digest facts to make business decisions. IT leaders need technically correct information to evolve their technology to meet core business objectives and effectively manage a dynamic environment.

So, lets get down to business. What do you need from me to achieve performance excellence?