dfox

16 ms·

151.
by dfox 2y ago
The reason why most standalone digital clocks use LED seven segment displays is that to this day almost all of them are built around the same weird PMOS Sanyo ASIC from late 1970's.
152.
by dfox 2y ago
Almost any other display that you will reasonably use in a application that makes sense for segmented multiplexed LEDs will integrate internal controller, so instead of having to refresh the thing somehow you have a thing that is simply an
153.
by dfox 2y ago
I always wondered how it is possible that MSCDEX is not an actual driver but mere TSR and now I know: CDs look like network shares to DOS. I suspect that there is some reasoning behind this “design” (different filename conventions on ISO-96
154.
by dfox 2y ago
Few years back the electronic lock on a safe in our office somehow died. The locksmith who supplied it was on vacation, but we somehow were able to convince him to give us the drill plan with the assurance that he will be the one to fix the
155.
by dfox 2y ago
If you are serious about printing labels and can program you should get yourself a Zebra label printer or Intermec/Honeywell one that has ZPL emulation (these tend to be cheaper than Zebra, so near the pricing of Brother consumer stuff
156.
by dfox 2y ago
Since the start of using any kind of SMS 2FA I keep noticing, that for some systems you get “nice to remember” numbers somewhat often. I didn't care enough to actually write them down and confirm whether it is a fact or just my interna
157.
by dfox 2y ago
At a large enough scale you can do most of these processes by automation, except the actual courier delivery. And well, at least UPS has actual identified locations in their vans, so just going by location+barcode might be somewhat practic
158.
by dfox 2y ago
Free of charge to you and free of charge to the sender. Which is exactly my point, for PostNL it is a sunk cost and if that will happen for significant amount of their consignments it would be an issue.
159.
by dfox 2y ago
The availability of centralized API was not the issue. The issue was that the operators were significantly more efficient when performing the process “by eye” than when supported by some kind of computerized system. For example the on-depot
160.
by dfox 2y ago
When we started our own parcel service in 2015 we thought that everything that is needed on the actual parcel is the unique ID of the parcel (generated by a system reminiscent of twitter's snowflake and intentionally printed with digit
161.
by dfox 2y ago
For a mail/parcel carrier relabeling anything is (somewhat surprisingly) a significant added cost (and there is not that much of margin), so they will do everything in their power to not do that unless absolutely necessary.
162.
by dfox 2y ago
Well I remember one project where having the app small and portable or at least self-contained was a hard requirement. It involved an embedded device (ridiculously complex and over-engineered embedded device) that used customized USB flash
163.
by dfox 2y ago
The issue is not really about the verifiability of the bytecode but about the interpreter checking the invariants. JVM bytecode is intentionally designed to be verifiable so that the inner loop of the interpreter does not need to care about
164.
by dfox 2y ago
This idea of accepting some kind of subset is exactly what JVM does. There is a set of rules (IIRC 29 of them) that the JVM bytecode have to follow to be accepted, the one important rule is “stack entries should always be used as a same typ
165.
by dfox 2y ago
Unreal Engine has something that can be called embedded web browser since the day one. The original Unreal Engine is this thing that has its own implementation of “something not entirely unlike JVM” and refers to various things by means of
166.
by dfox 2y ago
The hardware is not capable of conjuring up new characters, so this is limited to detecting things like bold and underline… and maybe mapping the overprinting to characters that already exist in the font.
167.
by dfox 2y ago
I vaguely remember that vt220 actually special caces this kind of overprinting. And today everybody targets some kind of vt220/xterm hybrid…
168.
by dfox 2y ago
Probably the most common reason for the integrated wifi card becoming “non-deterministically non-recognized” is that the OEM fucked up the rf-kill in that there are slight differences between what the EC firmware and ACPI expect and how th
169.
by dfox 2y ago
Same thing will happen on macOS with SharePoint/OneDrive. Found that out the hard way last week.
170.
by dfox 2y ago
It is my understanding that what this paper describes was never actually shipped and instead the StarLisp was significantly lower level affair, on par of the other Star-something languages for the CM, or rather the frontend of it (CM is ess
171.
by dfox 2y ago
Twenty years ago (when building community wireless networks was all the craze, at least here), we played with free-space optics and in course of that found out that random soviet (4mm!) green LED from parts bin makes exceptionally good rece
172.
by dfox 2y ago
The inefficiency is about going through kernel that will then dispatch some signal and the signal handler has to analyze what exactly happened, that is not a slow path, but ridiculously slow path. OTOH, my view is somewhat LISP-centric and
173.
by dfox 2y ago
Actually SPARC have four instructions that are directly meant for efficient implementation of Lisp/Smalltalk that came from SOAR and SPUR. Using that on top of some kind of Unix is shall we say problematic (in same way that x86 BOUNDS
174.
by dfox 2y ago
In java lamdas are just a syntactic sugar for anonymous inner classes, with the same resulting limitations, so one can either argue that java always had lamdas, but with weird verbose syntax, or that java does not really have lambdas (in wh
175.
by dfox 2y ago
Then they found a way how to (ab)use amd64 MMU to do more or less the same thing that they had custom CPU architecture for (ie. GC barriers in hardware).
176.
by dfox 2y ago
That is mostly working system except the nss module might depend on stuff like environment variables or SELinux contexts. Also, when you introduce per-user services you are deep into the “reinventing systemd” territory. Interesting idea wou
177.
by dfox 2y ago
> That daemon should be set up by the system administrator. One of the reasons why nsswitch runs in the client process is that the modules may very well need access to files that are only readable by the user (kerberos tickets...) or in
178.
by dfox 2y ago
On other typical 32b MCUs you end up doing this kind of bitbanging purely from software, while RP2040 has the PIO which is kind of programmable bitbanging accelerator. Similar hardware blocks in other MCUs are either single-purpose (think U
179.
by dfox 2y ago
Pin 9 is there to power the EEPROM in the monitor fromthe computer even when the monitor is turned off, so monitors providing 5V there would lead to same kind of problems as there are with DisplayPort pin 20 and cheap cables that connect th
180.
by dfox 2y ago
I assume that design based on something like TPS92411 actually ends up being cheaper and more reliable than various series capacitor constructions, while being dimmable just fine and without spewing EMI. So it is just question of the OEMs h

More ›