Are you ready for the Leago Switchover?

We’re getting close to announcing the anticipated schedule for the Leago switchover.

TDs should start to become familiar with the new format for editing and validating events online on Leago. (Events can also be uploaded using the 3 DBF files.) So far it appears that only about 3 dozen TDs have experimented with the Leago system during the beta test.

Players, coaches and parents may want to become familiar with the way player and tournament data is searched and displayed on Leago.

The beta system can be used for both of these, see https://beta-ratings.uschess.org.

Right now, I can’t believe that I would be in a position to say that I would rather use the old system instead. I’d rather put up having to put in the “XX” to change section names, then use this new system. It has A LOT of potential, and a few feature that are improvements, but I have recommended a number of changes over the past several weeks and little has been implemented. I process around 300 tournaments every year, and this new system is going to make the task of uploading, validating and rating each of them take significantly longer!

tom

With the color validation added back I’d be willing to learn to live with it.

The color validation makes it easy for a TD to quickly review errors and anything else that might need attention. If I see a complete green screen, I am happy. If I see anything else I can check to see what is in my SwissSys or WinTD files and is what i see on the upload screen what I expect to see from SwissSys/WinTD.

I have never used the older system’s entry method for entering tournaments. I have used Leago to enter tournaments manually entering data and uploading the 3 files. Leago seems simple. However, if only about 36 TDs have used Leago to enter tournaments, we are not ready as a group to use Leago. Have all the bugs of Leago been fixed? Leago needs to be more topic driven. Having headings for sections of data entry are a useful idea so that people have a general idea what to do when they see the individual sections. Topic headings could consist of brief instructions of how to enter data instead of assuming the data entry operator knows exactly how to enter data.

Considering that over 1350 individuals were listed as having been the chief TD of an event rated in the last 12 months, the low number of beta testers is a big concern.

What’s needed, IMHO, is for hundreds of TDs to test the new system, so there’s an overwhelming expression of sentiment about its features and flaws. 2 or 3 requests/complaints might not get noticed, several hundred would be hard to ignore.

The switchover is going to happen when US Chess leaders feel confident that it will work. Feedback, especially from TDs, is likely to be a major factor in their confidence level.

To quote a movie line: You don’t go when you’re ready, you go when you’re ready enough.

“Ready enough” needs to include all essential functions that were determined ahead of testing plus those that were flagged during the beta testing. Again, it would be good if there were title headings which announced what each data entry section was about or some minute instructions on how and why to fill out the data entry sections of the Leago filing method. These instructions could include things like when a person needs to enter an Organizer and if the Organizer can be the same as a TD.

So many folks I poked about the new system over the last year seemed to have no idea there was a new system coming – Just last week I saw some new chess parents being educated on where to find their child’s info and how to look at the tournament results …. I was waiting for the “… and it is all about to change …” – so I mentioned it to the educators and they batted 0-4 on knowing things were about to change – 2 of the 4 do submit tournaments.

I do know two TDs who still insist on hand entering tournaments even though they are using WINTD for pairings (2-40 players 1-3 sections). I’ll guess they are not participating.

I still can’t get into the site with my chess (older hardware) system so I have not tested very much. They should have just ditched the dancing chess pieces by now and worried about the real fix at a later date if it is that difficult & worth doing at all.

Some “vendors/developers” just want to get done with a project so that they can move on :grin:

Since Leago will have an ongoing maintenance/support agreement and manage the Azure server, I don’t think that’s an issue here, we are anticipating a long and happy relationship with them.

Changes tend to occur in three stages. First, Leago does them on their dev system. Then those changes get moved over to the alpha server for external testing, and then finally over to the beta server. I think several changes and fixes are due to be rolled out to the beta server in the next few days.

Once we get to a production environment, there will probably be a similar dev-staging-production rollout cycle. We’ve talked about leaving the beta system in place as a ‘sandbox’ for TDs and software developers to utilize as a learning tool and for testing API-based access to the Leago system, for querying for results as well as for uploading events directly into Leago rather than through the 3 DBF files. (I don’t know what the monthly costs are to keep that server up, that may factor into that decision, or the staging system may also serve as the user/developer sandbox.)

I would like to say go ahead for switchover if US Chess has a rollback to go back to older system. If not, my suggestion is not doing switchover with current test situation. You will find more problems in production with real TDs using it then on this test environment, guaranteed. Problems can GO AWAY, for example if me just need to upload 1-2 yearly, because I will just find work around to enter manually. But for TD who enters 100+ file yearly, additional manual or problems will drive them crazy and sad.

The TD/A system is definitely our fallback option. We will turn TD access to TD/A off at the start of the migration process, but it could be turned back on at any point, and we will continue to keep the membership data on it current as there are some other tasks running on the legacy server (Gambit) that haven’t been scheduled for migration yet, like correspondence chess.

The fallback option would likely mean that any events that had been entered on Leago would need to be redone on TD/A.

I haven’t fully mastered the data structure Leago is using yet, it is very different from the one we’ve been using for the last 20 years, and in some ways since 1991. Would it be possible to back-port new rated events from Leago back to the legacy system? Probably, but we’d also have to make sure any changes to other rated events are also back-ported, so it’s not a simple solution. (We process dozens of changes to previously rated events every week.)

There was some discussion of running a full-blown parallel test, but since that would require 1300+ TDs to do all the work to enter all their events twice, that was pretty much a non-starter idea.

The office plans to send out another message to all TDs and affiliates urging them to use the beta system to get familiar with the Leago interface, because lot of things are done differently.

I find it takes a few events to build up speed in both reviewing validation issues and entering events online. A lot of that is just knowing where everything is.

The plan is to refresh the database on the beta system on Tuesday or Wednesday, which would wipe out any pending events in your workarea, but you could reload them.

The updated database, along with deploying a number of bugfixes and code changes, should resolve a number of the issues that have been reported, both by the beta testers and by staff.

1 Like

A perfect example for me is that I just tried uploading my weekly blitz tournament to the new system, and got “Error:” with no details. No contact, no references on what’s wrong, no hints on something being amiss, no idea what-so-ever “wasn’t right” with the DBF files. They uploaded just fine with the legacy system less than 5 minutes later, when I got fed up with trying in vain. Perhaps it’s just me - but after running a tournament, the last thing I think about doing is beta testing. The slightest friction like this - makes me want to do it even less. I’ve tried other files that I’ve uploaded successfully (Vega Chess)… similar, no warning, no nothing - just a useless “Error:”… hard to test a workflow in a system that doesn’t let you even do the basics, or tell you what’s wrong. Devil’s advocate to others - if something like this moves to production, and I can’t upload a simple tournament, and have no ideas what’s wrong - I’m just going to stop doing rated tournaments, rather than fight the system or enter them by hand - I (like a lot of club TD’s, I’d assume) are losing money (or at best breaking even) on events.

Thanks for taking time to participate in the beta test.

I understand your frustration completely. I spent a half hour last night trying to get an event through validation so I could check that the FIDE pricing on it was working, and my thoughts at the time were that if I was just a TD I would have probably given up after about 5 minutes. (It hasn’t been helping that my home Internet connection, a 1GB fiber connection, has been unstable recently, kicking me off remote US Chess sites I’m logged in to, etc.)

I’m assuming this is your event: Pawn to be Wild Event ID 202510075562

Looking at the DBF files uploaded for that event, I notice two things. One is that these are the original DBF file format, without color information. It is also possible that the files are slightly malformed, the ending appears to have an extra ‘00’ byte in it on all 3 files. It is possible that a tool to read DBF files that pays strict attention to the formatting would reject it as malformed. I can check with Leago on that in the morning.

FWIW, I tried uploading it as well and also got an error.

What program (and version number) created it? Can it create the Draft version 2C files with color information?

I think the refresh of the Leago database, which will probably happen tomorrow, will coincide with a deployment of several changes. I would ask you to try again in a couple of days.

FWIW, I was able to track the time it takes from when a file is uploaded until the TD has it ready for payment. (I log EVERYTHING!)

Looking at 388 recently rated events, 56% of them took under 15 minutes from start to finish and nearly 3/4 took under an hour. 10% took a day or longer, probably to resolve membership issues.

That’s the performance level we should be shooting for with Leago. We’re not there yet.

The lack of recent synch makes it impossible to actually try and submit any of our recent events. But I note that there is an alert that doesn’t understand that people might have more than 2 names or might have suffixes:

Leago says the update of the database on the beta system, which will wipe out all existing uploads, should happen later this afternoon.

This will be more current member data, from September 25th, I believe. I realize that’s still two weeks old but it is better than August 5th. :sigh:

Another step forward will come when Leago is able to pull membership data directly from Gambit. (We got the firewall configured to allow this yesterday afternoon.) That may enable us to do a much more current update of the member data on the beta system. (Gambit member data is updated twice a day, at around 4 AM and again at around 6:30 PM.)

Real-time processing of member records pushed directly from the CIVI-CRM system (with a lag time of maybe 5 minutes, roughly about the same as we currently see on MSA) is the step after that. Leago should also have the ability to do a ‘pull’ from CIVI-CRM if an ID is either not found or says it is not current.

It appears that prodding TDs to start familiarizing themselves with the Leago framework has had an effect, as of last night it appeared that over 50 TDs had attempted to either upload an event or enter one online.

I’ve got a bucket list of things to check after the DB refresh is done, hopefully it will be a big step forward for everybody.

It looks like the database reset on beta won’t happen until tomorrow. Alpha has been reset, so staff and possibly the alpha testers can do some testing there.