teotwaki

teotwaki, 18 days ago

@augieray @cstross I caught Covid (for the very first time) at the Stockholm concert.

teotwaki, 22 days ago

@jerzone Having used a few of these on a number of quadrupeds, they don’t need to be as long as you think. If they’re well designed, they should allow you to “shoot” the pill down the pet’s throat. As long as you are able to land it behind the base of the tongue, they can’t spit it back out provided you shut their mouth and massage the throat to trigger a swallow.

If it’s too long, there’s a greater risk of injuring the animal’s soft tissues. Fun project though!

teotwaki, 27 days ago

@bram @stooovie I’ve stored PLA and TPU like that for 5 years and it preserved perfectly. The TPU I finished printing this week.

This being said, I did invest in a PolyDryer and a few boxes as I’m moving to more difficult-to-handle filaments. Nylon is one that sounds quite bad, but very exciting.

teotwaki, 1 month ago

@DonCipriano @cults3d How do you insert a person’s head and their hands?

teotwaki, 1 month ago

@yanncphoto Salut! Je fais pas mal d’impression et de modélisation. Quelle est la taille des gonds? Et quelle force doivent-ils gérer?

teotwaki, 1 month ago

@TechConnectify This is very much in line with current events.

teotwaki, 1 month ago

@tudsamfa @masukomi There is one thing: once you reach level 10 you can pick a username without the random string of digits appended to it.

teotwaki, 1 month ago

@TechConnectify beautifully put. I wonder if this may also be a Mastodon UX issue. I don’t use other social networks so I don’t know if it’s a feature elsewhere, but I was surprised that I could follow hashtags. I remember on Twitter you could have a cursory look at a hashtag, but that was it. Here it becomes part of your timeline. So any time someone shits on a topic I love, I see it, and it’s right there.

Randall was right, yet again.

teotwaki, 1 month ago

@TechConnectify To be very honest sometimes I wonder how you stay sane. I get maybe 15-20 notifications a day on my phone, total, and that’s in periods where I’m socially active.

I had to mute a YouTube comment that was getting a few likes every few days.

How many do you get, and how do you handle the cognitive load? Do you have an army of interns handling your socials as your persona slowly gets digested by internet famedom?

teotwaki, 1 month ago

@TechConnectify I guess it makes sense that with the scale of YT and other platforms, they’ve had to setup these systems, otherwise everyone would’ve burned out by the end of 2007.

I’m sure for a lot of people here, the fact that they actually do get replies directly from you and directly to their ideas is a big reason why you’re getting so much engagement. Again, no idea whether this is something you do on other social networks, but I appreciate it anyhow.

teotwaki, 1 month ago

@TechConnectify Do people have a real expectation of fairness on SoMe? Everyone writing you is semi wracking their brains trying to outwit, outsmart, and out-edge the other commenters so you’ll spend your attention on them.

SoMe and YT comments sections are the reverse uno card for people with a decent following. You try to make content that’s engaging for your audience on your YT channel. The commenters are trying to be engaging for you.

teotwaki, 1 month ago

@TechConnectify (It’s nearly midnight and I’m walking the dog, ability to respond in a timely fashion is compromised)

I think you need to figure out what you’re optimising for. I’m sure your patrons would love to have these discussions in our stead, and one could argue it would be fair for them to get more of your attention.

How many comments do you get per video/post? At what point does the random vs weighted comments stop mattering? Maybe highly engaged comments are important corrections?

teotwaki, 1 month ago

@TechConnectify Some people get a kick out of seeing another channel they know as the top pinned comments. The silent majority loves knowing that the nut/error/bug they cleverly spotted was indeed correct and they were right.

Has YT as a whole not basically already moved past expecting the channel to individually reply to comments?

What do you want to optimise for?

teotwaki, 1 month ago

@TechConnectify not sure you saw the above message in the notification frenzy, but I’m genuinely curious in the answer.

You mention you don’t want to do community management, so what are you trying to get out of it? Do you want to reward some viewers?

And are the different platforms comparable? Is it TC the brand/channel debating stuff on Masto or is it Alec enjoying joining the fray?

teotwaki, 1 month ago

@TechConnectify @jimcullen @VATVSLPR @vl_tone I’m not sure I’m fully onboard with that. I’m partial to the Slashdot philosophy: in a real conversation you can never unsay something. You can only add more to clarify your position.

If you, by blocking me, hide all your messages to me, how will I ever be able to understand your point?

Isn’t hiding the discussion from public view just gaslighting bystanders who may have see a glimpse but now can’t even read the beginning?

teotwaki, 1 month ago

@TechConnectify @SteveHeist @markwaypoint The one thing that bothers me is how easily people are called reply guys, and how being blocked by someone destroys the entire threading system, preventing you from participating in the rest of the conversation.

teotwaki, 2 months ago

@mackuba 10s or 100s GB are easy pickings for either one—the whole dataset can be held in memory at any given point. Make sure you aggressively review indexes, and index types on a regular basis. If you have high writes vs reads, maybe reconsider why you have a relational database to start with? Would anything else fit your needs better? Make sure you understand the clustering strategy, test backups, and test assumptions from your cloud provider.

teotwaki, 2 months ago

@mackuba So it sounds like it doesn’t even need to be in a database, and is mostly (exclusively?) append-only?

In that case, why not use Parquet files and not use a server at all? Store the files in S3, query through Athena if needed, do exploratory analysis through pandas, partitioned by day so you can still append your data every day, etc.

It’s more data sciency rather than RDBMS, but it’s a workflow used by thousands of data scientists.

teotwaki, 2 months ago

@mackuba In that case, just stick with MySQL, considering you’re most familiar with it. You don’t need clustering/failover, backups can be handled manually, and don’t setup any indexes beyond PKs or something. Queries will be slow, but that should be fine for your use case.

If you can afford it, get enough RAM to host the entire dataset. It makes querying a lot faster ^^.

teotwaki, 2 months ago

@mackuba And I don’t mean that in a “well if you don’t want to learn anything”, but rather “well this is going to suck regardless of which RDBMS you throw at it”.

But just in case I can convince you to use the right technology: Parquet files are column-oriented files that can be used to store TBs. S3 can store unlimited files, and Athena gives you SQL-like access to the data in the files, where each Parquet file is a table.

teotwaki, 2 months ago

@mackuba Pandas and dataframes are ways of representing data operations on top of columnar data (such as Parquet files). This allows you to do the same kind of operations as SQL, joins, selects, etc, but without ever leaving the Python REPL. It’s easily scriptable and insanely fast—considering.

I know of companies that store TBs per day like this.

Good luck in any case with your project!

teotwaki, 2 months ago

@shanselman I’m not sure why people consider this to be the work of a single person. I remember reading that they only connected through a VPN. If this was indeed an orchestrated state-sponsored attack, I would more readily believe this was a group of 4-5 people, with a dev or two, a writer, a coordinator, etc. The bus factor would be too high otherwise.

This guy appeared in the middle of the Covid pandemic, to exploit highly stressed maintainers and social protection systems. No coincidence.

teotwaki, 2 months ago

@billy Why are dependencies bad? You’re discarding an entire language just because it uses dependencies?

Any decently sized Python project will pull in a hundred dependencies as well. Same for Go.

teotwaki, 2 months ago

@bmaxv @thephpf Where did you see that they were getting paid? I understood it more as an advisory role to help fashion regulatory, which is sorely needed.