2,545 karma · joined September 26, 2014
It's rare enough that the CEO publishes a blog post every time they do it.
"This is a hard case and we would caution anyone from seeing it as setting precedent."
Higher screen resolution uses more memory too.
I'm sure dependencies / libraries have grown in size too, and yeah, using web for UI takes memory.
https://en.wikipedia.org/wiki/Dead_Sea_Scrolls#/media/File:G...
It often can look better to have more whitespace, but it lowers information density. It depends on the situation.
Browsers used to default to adding margin/padding automatically to documents (Quirks Mode), but modern html includes a doctype to use "Standards Mode" which is more sane.
At the same time it would be nice to default to sans-serif typeface and line-height around 1.5 but it starts getting arbitrary very quickly and it's nice to build on something that doesn't change over time.
Minimizing the number of dependencies helps a lot too.
But don't blindly npm or pip install something unless you trust the developers. npx/pipx are even worse. All it takes is a one typo-squatter to steal your ssh keys and maybe even saved browser passwords or cookies.
You just have to manually call json.parse(request.body) for incoming requests and return JsonResponse(data). And you generally need to serialize the data yourself.
Django does have some built-in model serialization, but that's mostly for import/export data, not really for sending to the browser.
https://www.reddit.com/r/radiohead/comments/mhpv3/we_need_to...
(You can hear this clearly if you isolate stereo differences with Audacity: Effect -> Vocal Reduction and Isolation -> Remove Center. Strength 100%.)
Yes, if you do WebSockets, async really is a must if you don't want to have one thread per active connection. Each thread takes 8mb of virtual memory, not sure how much actual memory.
It's also nice for a backend that's doing a lot of proxying apis to another backend.
"There is a problem with using the sys.exit() in a forked process (child processes or workers) instead of using the os._exit() that is mentioned in python documentation."
https://github.com/benoitc/gunicorn/pull/2820
I've always just needed to manually reload the gunicorn async server in development. Super annoying.
Basically every function call that does db-queries needs an "a"-prefixed version. aget(), acreate(), aget_or_create(), acount(), aexists(), aiterator(), __aiter__, etc. And in future versions they'll work on the lower level, undocumented api, like _afetch_all(), aprefetch_related_objects(), compiler.aexecute_sql(), etc.
Eventually you'll probably need to switch database drivers to something that actually supports async.
It makes sense as backend for a native iOS/Android app, and if you already have that REST endpoint then it can make sense to re-use it for the web-version of your app.
But if you're not doing an iOS/Android app (most websites?), then I generally agree it's a lot more work compared to form-based / server-side-templates (though sometimes can give a _slightly_ better user experience, _if_ done correctly.)
HTMX seems to be growing as a "REST endpoint" alternative, but again not as useful if you're also doing a native iOS/Android app.
It allows callbacks from numbers you’ve called recently which is nice.
For texts there’s “Filter Unknown Senders”
A, B, C, and D,
Pray, playmates, agree.
E, F, and G,
Well, so it shall be.
J, K, and L,
In peace we will dwell.
M, N, and O,
To play let us go.
P, Q, R, and S,
Love may we possess.
W, X, and Y,
Will not quarrel or die.
Z, and ampersand,
Go to school at command.
- A Book of Nursery Rhymes (1901)https://en.wikisource.org/wiki/A_Book_of_Nursery_Rhymes/Part...
_Any_ website can tell whether or not you have TeamViewer installed. Ad networks could theoretically target you based on whether or not you have TeamViewer installed.
I’m not assuming malice, but it’s a much bigger privacy hole than just increasing fingerprinting by a few bits.
The small-town post office will sometimes _delete_ the physical address in the mail system so the actualy physical address is completely in-valid (or worse gets autocorrected to an address a mile away).
Turns out FedEx and UPS base their address validation off USPS and also don’t ship to PO Boxes.
- Keystrokes and edits are instant (not just predicted), even on a high-latency internet connection. (Since ~2007 I've been constantly annoyed by what's called "bufferbloat", and I want to be more resilient to a not-perfect internet connection because I don't control internet latency at coffee shops, etc.) (Mosh is annoying, having to set up special firewall rules to allow UDP ports, etc, and doesn't fully solve the problem.)
- VS Code is also easier to pick up where I left off if I lose the connection. (Yes, can be mostly solved with screen/tmux/byobu, but annoying to manage that.)
- Nice typing support, autocomplete and code navigation. I can command+click functions to jump straight to their actual definitions (not just something named the same thing). I get more instant feedback, catching more errors before actually running the code. (I use Python + pyright/pylance, and I suppose vim has some plugins for this too, but it's nice having something officially supported by the editor. I'm also guessing VS Code is way more powerful.)
- Easer to deal with multiple open files and switching between files in a projects. Uses many of the same keyboard-shortcuts as browser tabs. (I suppose vim has this functionality, but I never learned it.) Project-wide find/replace is also really really nice. (Easy to toggle between regex and case-sensitive modes, etc. Easier than "pie": perl -p -i -e 's/this/that/g' $(git ls-files))
- I can drag local files to the remote server really easily, and I can also right-click->download any file I see.
- Files open in the editor automatically update contents on screen if their content changes on the file system (unless there are unsaved changes).
Downsides so far are:
- VS Code is pretty memory hungry on the server and uses multiple node.js processes that probably don't share a lot of memory despite sharing tons of dependencies. (So, yes, bloat.)
- Remote SSH plugin isn't open source, and kinda just downloads and runs closed-source code on the server. I trust Microsoft more than some open-source projects to not abuse this though ideally I'd prefer to apt install something that Debian or Ubuntu has vetted.
- Annoying having to install a local client. Maybe we'll get a fully-featured browser-only version of it someday, but from what I've gathered it's not there yet (probably harder to lock down security-wise too).