Fencing websites' availability report for July

[brtech]

I'm sorry, but this problem is more than a year old, and if he can't fix it in a year, then we must have another answer. Right now, I get the feeling that the plan is "the big fix" = a complete rewrite with no old bugs and all new bugs. Since the "big fix" is coming Real Soon Now, no one wants to put real money in fixing the old problems.

This is a mistake. We should pay a professional programmer (not a web designer) to find and fix the problem. It MUST be someone intimately familiar with whatever tools were used to generate the site. We'll also probably have to pay Dan something to be there to answer the questions of the guy who does the work.

The "big fix" is sure to be a lot later than whatever the current estimate is
The "big fix" is sure to fall over on arrival, although we can hope it gets stable soon

I do not understand why the board cannot authorize the staff to address the problem RIGHT NOW.

If someone will name the tools, I will do the googling it takes to find someone to fix the problem next week.

[Slim]

It makes one wonder how the USFA could deal with the management of a building (a la Tourny Task Force suggestion) when they cant even manage a website.

[mfp]

Brad,
In what way the message needs to be articulated to the USFA? It is simple and clear that the website is important to the membership and that USFA is not doing enough to satisfy the needs of the membership.

The membership wants:

• to have the website on >99%+ of the time,
• to register online for national tournaments,
• to see more detailed results of competitions,
• to renew membership online

Also the membership might want a website that performs acceptably when it's actually up and available. Monitoring shows that there were a significant number of hourly periods in July when the site was "available" however the average response time for requests took over 10,000 or even 15,000 ms.

Those UFSA website response time "health" values were after subtracting out the time measured it took to get a response from a reference website (google).

[brtech]

It makes one wonder how the USFA could deal with the management of a building (a la Tourny Task Force suggestion) when they cant even manage a website.

Having talked to Michael about it, I actually understand this a little.

The problem is that the staff has no expertise, they have a large number of people supplying advice for free, and they really don't know how to proceed. Michael clearly understands that most of the free advice he is getting is worth every penny he is paying, but doesn't really know how to separate the good advice from the, uh, less informed? advice. They've made some mistakes in dealing with the website, and they mostly get slammed no matter what they do, at least that is what it looks like from their point of view.

Still, I think inaction is the worst course of action.

[occasionalfencer]

The membership wants:

• to have the website on >99%+ of the time,
• to register online for national tournaments,
• to see more detailed results of competitions,
• to renew membership online

JEC[/QUOTE]

Also that whatever information is on the site is correct and up to date. For instance, the website still lists Paul Soter as the National Men's Epee Coach. Back in April, one member of the USFA reproached me for not realizing who the new ME coach was. I still don't know how I am supposed to know these things if the "official" website does not have the information. Does the USFA depend on gossip to spread information? I would much rather that there was no information in a given area than that the "information" is incorrect.

[mrbiggs]

* The uptime problems have been attributed (by DK) to an undiagnosed bug (or bugs) in the coding of the site and it has been explicitly stated that it is the site-specific programming that is crashing rather than the server, which remains up. The hosting is paid for by the USFA, maintainance is not, but is rather done on a volunteer basis by DK (AFAIK). Because of this tracking the bug(s) takes a lower priority than regular paying work and the problem remains unsolved.

Could we just bring in someone from the outside as a one-time sort of deal and just have them solve this bug? It seems to me that it would be worth the money to have a working website.