|
I think, it's roughly the same for me. I surely got a nook or cranny I'd like to polish for a bit before I think about running this thing on a rented server (main raison d'etre for WSL-debugging in this particular case, rented Windows servers cost an order of magnitude or so more, than their Linux counterparts).
Come to think of it, the vast majority of my code is OS-agnostic so I could, actually, start porting this thing more or less right now.
Thanks for inspiring me for taking the next step in this project!
|
|
|
|
|
You're welcome!
Mine is a desktop GUI... so using Desktop / OS tech.... But hey, I am curious about Avalonia!
|
|
|
|
|
Mine is, and I'm damn grateful I don't get to sink my teeth into the thorny topic of Linux GUIs, a server-sided console project.
To be honest, I've no idea how I would go about programming something GUI for Linux. I mean, I got an emergency plan should I ever get product requirements about Linux support for a GUI project at my workplace and that emergency plan would be Windows.Forms under Mono. Still, I hope https://github.com/jsuarezruiz/forms-gtk-progress/issues/31 to yield some results in the years to come.
WSLg should at least provide a sensible debugging experience in this scenario, it can run Edge after all, so it may just as well run my GUI.
|
|
|
|
|
It's called Windows Subsystem for Linux, and yes, it's production stable and in the regular build.
I recommend WSL 1. It doesn't need a hypervisor, has seamless file-system access, and you can just pick a distro you like and it works without X and without deamons. It's like running Linux natively, but much better in every way, since your drivers actually work and Windows Defender is lightyears ahead of anything in penguin land.
WSL 2 adds UI and service layer support to WSL 1, but uses Hyper-V to virtualize the file-system, which is stupid on multiple levels. The worst thing about Linux is X11/Wayland, the second worst thing are the deamons. Offering both and as a trade-off to being forced to use Hyper-V is like agreeing to get slapped in the face, so you can get punched in the gut.
When WSL1 is used right, you can integrate any Linux-specific toolchain on Windows. Right now I'm running Guacamole, SQLPad and code-server with SSO via tunnel over cloudflare access, all running inside WSL1 with less latency than if I was using a native Linux distro.
|
|
|
|
|
Well.. my Linux interest was to test Avalonia, as a substitute to WPF on Linux & Mac.... so.. I need X11!
|
|
|
|
|
Super Lloyd wrote: is there that many normal non IT user using Linux out there?
When the disk failed on my wife's underpowered vista laptop I slapped Ubuntu on it and she loved it. I was much faster and was fine for everything she needed (browsing, writing a few documents). The only problem she had was sending / receiving documents in Microsoft word format - mainly because other people were not used to importing docs in other formats.
My wife is most definitely not a IT person.
|
|
|
|
|
Yeah, my wife wouldn't have a problem running Linux. Over the years I've migrated her to using cross platform tools (Firefox, Thunderbird and Libre Office). About the only thing I'll have to do is install Zoom and she'll be set. I figure when Win 10 support is done, I'll just install whatever my current favorite Linux is (currently MX Linux).
|
|
|
|
|
Program: Will not work for anything...
Programmer: Doesn't know why...
Program: Works 100%
Programmer: Doesn't know why...
You're welcome...
|
|
|
|
|
Just had one of these moments with SQL Server. Bounced the server but I couldn't find out why one table in one database was locked.
|
|
|
|
|
All the time...
Also the classic stages of dealing with a support case:
1) Before attempting to reproduce: It is working fine, we would have caught it if it is broken.
2) Before troubleshooting after reproducing the bug: How is this broken?
3) After finding the issue: How did this ever work?
4) After checking in the fix: This could never have worked!
|
|
|
|
|
oofalladeez343 wrote: Program: Will not work for anything...
Programmer: Doesn't know why... Why get mad? The manager and the owner of the company don't know why either. That's why I am paid in the first place. I have no idea today, but gimme some time and I'll at least isolate what is causing it.
oofalladeez343 wrote: Program: Works 100%
Programmer: Doesn't know why... It never works 100%, and if you don't know why then some of your code comes from StackOverflow.
oofalladeez343 wrote: You're welcome... Anytime.
Bastard Programmer from Hell
"If you just follow the bacon Eddy, wherever it leads you, then you won't have to think about politics." -- Some Bell.
|
|
|
|
|
Nope.
However, often the program has very nasty bugs. It doesn't work as expected and I don't know why.
Quote: Program: Works 100% That would be a miracle. Only God blessed programs and trivial ones could work 100%.
However sometimes I wonder how can it work for a quite amount of time without crashing.
"In testa che avete, Signor di Ceprano?"
-- Rigoletto
|
|
|
|
|
Well... I mean compared to how it wasn't working at all...
No code works 100% unless it is a simple HelloWorld code.
|
|
|
|
|
Messing around with flex/bison--have it working in a simple case. Moved to another application, same compile/link/etc.--won't link.
I seriously believe that Babylonia failed because they were the first to develop computers.
|
|
|
|
|
I was messing around with Python Turtle because I had nothing better to do...
And while I was messing around, I did this...
import turtle
def main():
t = turtle.Turtle()
t.shape("turtle")
t.speed(100)
t.right(10)
length = 1
t.color("crimson")
for i in range(250):
t.forward(length)
t.right(100)
length = length + 0.5
main()
And it looks very much like the golden ratio in the center...
So I tried this...
import turtle
def main():
t = turtle.Turtle()
t.shape("turtle")
t.speed(100)
t.right(10)
length1 = 1
t.color("crimson")
for i in range(250):
t.forward(length1)
t.right(100)
length1 = length1 + 0.5
t.penup()
t.setpos(0,0)
t.pendown
length2 = 1
for i in range(250):
t.forward(length2)
t.left(100)
length2 = length2 + 0.5
main()
And there was only one difference in the design,
where the code stops is mirrored... That's it... No other variation...
Any opinions?...
|
|
|
|
|
I was just reading up on the UltraRam from yesterdays news letter
UltraRAM Breakthrough Brings New Memory and Storage Tech to Silicon | Tom's Hardware[^]
It sounds awesome for computers in general, but it would have a huge impact on embedded applications.
feel free to chime in honey the codewitch.
to have a small embedded device that does not have to maintain power to RAM while waiting for a trigger would be a huge power savings for the embedded CPU. and if the device looses power all together, you wouldn't have to worry about writing data to flash or EEPROM, before loss. even to just have a 128K of it on an embedded CPU would be a huge step forward.
imagine having a device that someone powers up in 999 years after being dug up from some old found landfill and having it start working, is the stuff of sci-fi books. (I would say boot up, but if everything is still in memory, not much booting needs to happen)
|
|
|
|
|
If Scotty had had this, he wouldn't have needed to put the transporter buffer into a continuous diagnostic loop.
STNG "Relics" episode[^]
|
|
|
|
|
the Dyson Sphere episode, that was one of my favs
|
|
|
|
|
I loved that episode.
I loved when Scotty gave LaForge some good advice.
SCOTT: Do you mind a little advice? Starfleet captains are like children. They want everything right now and they want it their way, but the secret is to give them only what they need, not what they want.
LAFORGE: Yeah, well I told the Captain I'd have this analysis done in an hour.
SCOTT: How long will it really take?
LAFORGE: An hour.
SCOTT: You didn't tell him how long it would really take, did you?
LAFORGE: Of course I did.
SCOTT: Oh, laddie, you've got a lot to learn if you want people to think of you as a miracle worker.
Kelly Herald
Software Developer
|
|
|
|
|
Doesn't that also apply to us developers and the stakeholders?
|
|
|
|
|
I just re-watched ST:NG (all 7 seasons) last fall. I liked that episode, but one thing caught my eye.
If you were orbiting a Dyson sphere at anything less than a distance of several million miles, it would look like a flat surface. You couldn't see curvature.
Pedantic and fatuous, I know .
Software Zen: delete this;
|
|
|
|
|
This made me think, what would a Dyson sphere look like in the flat earth model, then I remembered they don't have one...
|
|
|
|
|
Well,
Just ignore the journalism, there is nothing to be excited about. The paper is theoretical and not a single piece of UltraRAM exists. Looks complicated to manufacture to me. Keep in mind that FeRAM[^], MRAM[^] and ReRAM[^] were all supposedly going to be used for storage and replace RAM. Never happened.
Best Wishes,
-David Delaune
|
|
|
|
|
|
I need to read more about this as it sounds almost like existing Pseudo-Static RAM (PSRAM) or maybe SRAM too? (I can't remember) technology that's been around for awhile the way you describe it.
EDIT: Now that I've read it, it works differently than PSRAM, and PSRAM despite being non-volatile in a sense, isn't usually set up to remember on powerdown. The non-volatility is more for avoiding needing complicated refresh circuitry in a memory controller. However, unless I miss my guess, it *could* be non-volatile on powerdown, it's just not used that way typically. I might be wrong there. It's an educated guess.
The upshot seems to be PSRAM style simplicity with DRAM style access speeds. Price is an issue according to the article. I wonder about that, since the whole point of using this stuff vs. traditional RAM is ultimately to cut down on cost (and perhaps power consumption)
This is all just initial thoughts and I'm fuzzy on some of the details of SRAM and PSRAM, since I've never needed to know them. So take all of this with a whole lot of salt. In fact, check with your doctor about your sodium intake before consuming this post.
Real programmers use butterflies
|
|
|
|