Blog/January 2017/Jan 6th
From RonWareWiki
< Blog | January 2017
January 6th
It was “back to work” for Esther this week, after a relaxing two-week vacation. She was enjoying her well-deserved break and not really looking forward to its end. But they like her at work, so she returned.
She was uplifted somewhat by the exchange of our dying leased car for a shiny new one. This one is a metallic-black Renault Clio (the previous was a Renault Fluence). It’s a lot smaller and more fuel–efficient (which is very good), and it seems zippier than the other (not necessarily good, but Esther likes that). May she enjoy it in good health!
We met up this week with friends of ours from the Olde Country, who we haven’t seen since making aliyah (almost) ten years ago. Actually, I saw them when I went back to visit a while back, but Esther hadn’t seen them since 2007. We had a nice long lunch with them, and got caught up with each other.
Sarah had a busy week as well, getting her sales-woman groove on. She also had an interview for an occasional job doing video editing, and the interviewer seemed to be reasonably happy — so she may have a second job soon. Welcome to Israel!
I managed to get my main client’s system put back together and working nicely. It was a fairly large bit of digital-surgery I had to perform, but the patient is doing well. I’ve also been working on the next release of 8th, and fielding questions from various customers. We just completed an ‘update cycle’ for our customers, and so far have about 70% retention — which is really excellent for a new product. So all’s good!
<techie stuff>
Dear techie friends, I have a SQL question (non-techies, feel free to skip this paragraph and the next one): I’m working on solving a ‘fair queueing’ problem. I’ve got an in-memory database (SQLite) which has a table of “clients” and a table of “files”. Each file may appear on multiple clients. I want to distribute jobs to the clients based on (1) the availability of files to them and (2) on their general availability. This isn’t such a tough problem to solve, and I did so by reading the information from the SQL database into C++ structures and doing things the old-fashioned way. But it irks me to copy information around if I don’t have to.
My question is: “how would you accomplish the distribution using SQL queries?”. I know I could simply use multiple SELECT/INSERT statements, with triggers on the INSERT to update a weight value for each client. Since I would have to invoke the SQL from my C++ code, I’m not sure that would be better than just doing it all in C++. It might, however, be easier to code and debug. I’m open to suggestions…
</techie stuff>
About stuff:
- US is doubling-down on agitprop against the Russians.
- What’s up with all the anti-Israel sentiment?
- You know you want to read Dave Barry’s year in review!
This shabbat is my birthday on the Gregorian calendar (you have until the 2nd of Shevat [Jan 29th] for my Hebrew birthday — I accept cash donations until then), and we’ll be joined by RivQua. On the menu:
homemade ḥalla,
chicken schnitzel,
oven-fried potatoes,
stuffed eggplant,
veggie chili,
noodles,
various salatim,
and Esther’s famous chocolate birthday cake. My preciousssss!
Until next time,
shabbat shalom!
Top: Blog | Prev: Dec 30th | Next: Jan 13th |