3,175 karma · joined October 22, 2012
1. This is a familiar problem and it is where I spend the majority of time. If a bank provides a unique transaction ID the it would be a simple matter of code; however, none of mine do. What I do when I spot problems is run `hledger areg <account> --begin=<last month> --end=<this month>` and compare with the CSV. Opening the output of that command + this months CSV in vim using vertical split (`vsp`) and scrollbind (`scrollbind`) so each line scrolls together works well and is pretty quick. It's still a problem and a pain.
2. This is solved in my system in a robust way (I believe) with a "virtual account." I do most of my spending on my credit card. Then every month I pay off the card from my checking. To represent this I need to see:
- liabilities:credit with any balance I carry
- expenses should show a transfer from liabilities:credit to the correct expense category/account
- I should avoid being "double charged"; i.e., when I pay off liabilities:credit it should NOT be an expense.
I've set it up like this:
- rules/debit/journal.rules: if (payment to credit) then account2 assets:transfer:credit
- rules/credit/journal.rules: if (payment from debit) then account2 assets:transfer:credit
This becomes a way to spot problems. If assets:transfer:credit ever has a balance then something may be awry and you should spend some time digging. I do the same thing for Amazon. Each Amazon order is for multiple expenses. I have assets:transfer:amazon. This is how I learned the the $0.29 Colorado retail delivery fee is not universally applied to orders from Amazon—that was a frustrating journey!
3. I haven't worried about this. Looks like other good responses in this thread.
[0]: <http://www.findu.com/>
weewx[1] software supports sending data to CWOP and has drivers for the Davis station among others.
Bonus: you can also send data to CWOP over the ham bands via aprs, and all your data shows up on aprs.fi[2]
[0]: <https://www.wxforum.net/>
[1]: <https://weewx.com/>
[2]: <https://aprs.fi/>
[0]: <https://blog.benjojo.co.uk/post/ssh-randomart-how-does-it-wo...>
Since then, by default ssh shows the SHA256 fingerprint in base64 (minus trailing `=` padding), rather than an MD5 fingerprint in hex. I probably shouldn't have elided this detail since it's confusing.
Titles are hard.
And the term "product bets" makes the leadership team sound like Mortimer and Randolph Duke.
Radio will continue to be a technology used for providing infrastructure where none exists (disaster areas, offgrid communications). And sometimes it takes the place of "last mile" ISPs to get people in rural access to the wider internet.
Plus, playing radio is fun.
On the First Amendment vs. licensing requirements: Most licensing requirements are aimed at curbing harmful interference. The only thing in Part 97 (I can remember offhand anyway) that limits what you can say is 97.113(a)(4)[0]—most of what it mentions is covered by other laws (obscenity and copyright). I think there's something about selling stuff, too. But hams are pretty much self-policing this, afaict.
Encrypted communication also prohibited (boo). But now we have meshtastic[1] on ISM bands for that!
[0]: <https:/ / www.ecfr.gov/ current/ title-47/ part-97#p-97.113(a)(4)> [1]: <https://meshtastic.org/>
It is a nice upgrade. But it was still difficult to justify the upgrade to myself—spending money on laptops is never something I relish.
And the X230/220 were both chugging along—they struggled with the modern internet more than day-to-day software tasks.
I'm happy to learn this as "Framework 13 AMD" is nicer to write than "Framework 13, 13th generation."
1. Something you have 2. Something you know
For the web, "something we know" is a password, so we've implemented something you have (e.g., a shared secret stored on a physical device).
For cars, "something you have" is a key. This system seems to implement the "something you know" part via entering a startup sequence (afaict).
Buy a cheapo handy-talky. Use radioreference to find your local repeaters. And get on the air!
You are required to announce your call sign. And you can find anyone by their call sign in the FCC's Universal License System.
Git is storing at least two copies of files: one in your working tree and one as a zlib compressed blob in the object store.
It may also store more copies if, for example, you updated embedded metadata and then have to recommit the file.
Zuul, of course, had this back in 2012.[1]
2023: GitHub catching up!
[0]: <https://graydon2.dreamwidth.org/1597.html> [1]: <https://zuul-ci.org/>
Wikimedia opts to close PRs with a message saying (tl;dr): this is a read-only mirror, contribute via Gerrit.
Importing from GitHub seems like it might be a bad experience except for drive-by commits (which are valuable and Wikimedia misses out on).
Mirroring to GitHub is sensible—(1) Dependabot (2) Nice code browser (3) this is a perk of working on open projects: your name shows up in GitHub (the default place for open source developers) next to the project.
But there's no trivial way to make it clear for developers that your repo is a mirror (outside of the description). It would be nice if there were a button in the UI to make it clear a mirror is a mirror (e.g., to show up in searches for `mirror:true`[1]). It seems like this requires Special Privileges™ somewhere...
[0]: <https://phabricator.wikimedia.org/T37497> [1]: <https://github.com/search?q=mirror%3Atrue&type=repositories>
My service has been rock solid: https://photos.tylercipriani.com/2023-05-25_next-light.png
I think haskell moved to GitLab, now Blender's on Gitea.
I'm curious if people are moving to phorge?
Wikimedia still uses phab for bug tracking. Taking a look at phorge Soon™
An additional resource I would add to it is Steven Pinker's "Sense of Style: The Thinking Person’s Guide to Writing in the 21st Century"[0].
Pinker's book advocates for the same classical style as Williams and Thomas & Turner. But Pinker also offers a deep dive into grammar, linguistics, and word choice (Pinker is a member of the American Heritage Dictionary's Usage Panel[1])
One other note: Joseph Williams's "Style: Lessons in Clarity and Grace" is available in roughly infinite editions under similar titles. PDFs of some of those titles are available for free online.
[0]: <https://stevenpinker.com/publications/sense-style-thinking-p...> [1]: <https://www.ahdictionary.com/word/usagepanel.html>
[0]: <https://www.abandonedspaces.com/parks/joyland-amusement-park...>
Why not? What should you do instead?
Message shows up in the chat room on the phone and on the other devices linked to the room. Planning to play with them for camping Soon™.
[0]: < https://www.aliexpress.com/item/2255800992363816.html?spm=a2...> [1]: <https://www.18650batterystore.com/products/sanyo-ncr18650ga>
Below is the email I sent to one of the authors 2021-06-07:
I tried to email this question to dora-data@google.com
and got a bounce back, so I figured I'd try this address.
tl;dr: I'm currently re-reading Accelerate in our book club
at work. Is there a copy of the survey available somewhere?
I'm curious about the questions asked. The book (and report)
say things like, "[Elite performers have] 208 times more
frequent code deployments" (state of devops 2019 pg 21)
and "[High performers have] 46 times more frequent code
deployments" (Accelerate pg 16).
The book also says, "We asked survey respondents how often
their organization deploys code [...] offering the following
options" with a list of 6 options like "between once per hour
and once per day".
How do you get from multiple-choice to assertions of
deploying 46 or 208 times more frequently? Is there
a copy of the survey available to get an idea
of the questions you asked?