Freezing Saddles Technology – 2020 Thread

[obscurerichard]

For the 2nd year running, I’m declaring an official Freezing Saddles Technology thread.

This year we’re again managing the Freezing Saddles software (hosted on GitHub) through GitHub issues and a GitHub Project Kanban board. I plan on communicating with the technology team primarily through a Freezing Saddles Slack chat system. I’m leading the team this year again and have been doing most of the programming and almost all of the operations, but I’m eager to teach others what they need to know. I’m hoping that @jrenaut will continue to help as he has in previous years (again! thank you!) and I’m looking forward to getting more active contributions from potential technology team members.

With that in mind, I’d like to call for volunteers, please post here if you want to contribute, even if you are not sure you have either the skills or the time to contribute.

I’d like to host a series of scheduled drop-in Freezing Saddles development sessions in Arlington and DC and (gasp) maybe even Maryland if we have any tech volunteers who live in points north. Getting started with developing on this system can be a bit intimidating but I’d like to help ease that transition. Please pipe up if you think this would encourage you to contribute.

If you have a problem with the Freezing Saddles web site, you can post about it here, or add a note to the board or to the freezing-web issues.

Last year we made excellent progress in maintaining and extending the system, after some initial hiccups in transition. @hozn originally wrote software that runs the freezingsaddles.org site using Python, Flask, SQLAlchemy, Alembic, and some other frameworks, and last year we had our first full year of running through Docker. It is currently running on a Digital Ocean droplet that hozn is graciously footing the bill for, but we may complete a move away from that system to a new home this year.

The main challenges we are going to have this year:

* Altering the scoring so that the new declining-points system
* Getting more than just me, @jrenaut and @hozn to contribute
* Fixing a whole bunch of bugs we found last year
* Getting weather data back online after the demise of the Weather Underground API
* Getting distributed logging and more performance metrics back online
* Moving the system to a more robust hosting arrangement

Topic

[merlin]

We could summon the shadow proclamation to rule on reverse loops, but run the risk of a judgment that they rightly be subtracted from your total in accordance with all that is good and proper. Total milage is pretty solid; Strava does not appear to change the value (unless you crop a ride) and in normal usage rides would never be deleted from fs.org so I can personally, but with all respect and no intended slight, insult nor vituperation, only suggest human error. That is to say, I ran a full resync on your data and the sum remained the same.

mysql> select sum(distance) from rides where athlete_id = 19607766 ;
+

---

+
| sum(distance) |
+

---

+
| 2619.2199988365173 |
+

---

+
1 row in set (0.00 sec)

[merlin]

The data are keyed against your Strava athlete id (the number at the end of your strava url, https://www.strava.com/athletes/19607766) so name changes should not affect results; only switching to a new account.

[Lynn Scho]

Merlin! I don’t know who you are, but I appreciate your responses to me. Lynn

[Lynn Scho]

Thank you, thank you, thank you, Merlin.

[obscurerichard]

@Steve O 197294 wrote:

There’s something odd about the weekly chart. The blue spikes are team 12.
[ATTACH=CONFIG]20887[/ATTACH]

We’ve finally resolved this issue!

[obscurerichard]

I’m regretting that I did not manage to organize any in-person hackathons for Freezing Saddles Technology this year, but I’ve been buried under the for-profit work I’m on the hook for this season. I’m way behind on the book I’m writing (about Docker) as well… and now with COVID-19 it is not a good idea to do an end-of-season tech meetup.

There’s always next year!

[Author]

Replies