new linux standard

Discussions on more advanced topics such as monolithic vs micro-kernels, transactional memory models, and paging vs segmentation should go here. Use this forum to expand and improve the wiki!
User avatar
iansjack
Member
Member
Posts: 4828
Joined: Sat Mar 31, 2012 3:07 am
Location: Chichester, UK

Re: new linux standard

Post by iansjack »

They don't want you to run Linux programs. They are paying you to develop Windows executables.
And that’s why, in the unlikely event that you managed to bypass any restrictions, your first day in the job would be your last day.
kerravon
Member
Member
Posts: 332
Joined: Fri Nov 17, 2006 5:26 am

Re: new linux standard

Post by kerravon »

iansjack wrote: Mon Jul 07, 2025 6:27 am
They don't want you to run Linux programs. They are paying you to develop Windows executables.
And that’s why, in the unlikely event that you managed to bypass any restrictions, your first day in the job would be your last day.
Again - that's not true. Not universally for all employers in the world.

Even though the end result is that I need to produce a Windows executable, if I wish to bring along my Linux executables as a productivity aid, that may well be allowed. Again - you do not speak for every employer in the world.

E.g. at one job where I had to sign the usual "cover your @$$" documentation saying that I wouldn't install any "unauthorized software", my team leader directed me to download and install various software that I actually needed to do my job. That bit I remember clearly. I didn't need to run pre-canned executables of my own at that job for some reason, but I did want to download my source code so that I had some tools like hexdump available. I assume I asked my team leader if that was OK, because I do remember downloading some of my source code. I'm actually allowed to *write* that source code if I need to, so there's no technical difference between that and downloading it.

And you say "last day" on the job as if anyone really cares and is monitoring you to try to get you sacked. That's normally not the case. People do want to cover their own asses, but it would be unusual for someone to make an effort to try to "catch" you breaking a rule so that they can sack you. The expression "turn a blind eye" exists for a reason. People normally need to report to their immediate manager that they GOT THE JOB DONE. And so the person you report to is not likely to be too fussed if you say you would like a productivity tool - at least one that doesn't require a purchase order.

How many movies have you seen of corrupt cops? "Training Day"? And you think the IT industry is so strict that you can't download productivity tools or you get sacked on the spot? Every single employer in the world is that strict? What percentage of world employers have you worked for? Or did you extrapolate from a sample size of 1 imaginary employer?

Also note that sometimes I will be employed to work on a Linux (or whatever) system, but they use Windows to access those systems. And I may wish to run a productivity tool on Windows for some reason, so that's where I may wish to run my external x64 Linux executables, even though I'm employed to work on some unrelated system. I wish to bring the productivity tools from my home PC to my work PC. Separately I may wish to have those tools run on the Linux/mainframe/whatever, but I would need to be in a position to do a source compile in that case if it is non-x64.

It depends.

I can remember one place that wanted me to do some work on the weekend or something, and it was monitoring, so I would mostly be idle, and I offered to my manager to do the work for free (or something like that), so long as I could use the work machine for some non-work-related software development. He gave approval for that. Asses sufficiently covered. Even if asses weren't covered - I never heard about anyone giving a ****. Giving a **** about something like that requires actual work and also other issues.
kerravon
Member
Member
Posts: 332
Joined: Fri Nov 17, 2006 5:26 am

Re: new linux standard

Post by kerravon »

BTW, I should be careful not to give the impression that I'm busy breaking/bending all the rules because I don't care about rules.

Reality is that I'm actually the one sticking my neck out, and doing work - not to have someone sacked (I'm the one more likely at risk of being sacked for being anal-retentive) - but pointing out that something that is happening in production wouldn't survive audit if the auditors knew about this, and can we get a procedure change.

I don't think the audits of development machines are done by actual auditors. They're done by sysadmins, so it is down to whether the sysadmin is anal-retentive. And I don't think it is possible to audit a development machine anyway - we're allowed to write software - it's our job. So there are non-standard executables that can't be automatically detected as not following some rule. I assume the sysadmin is simply happy that it passed an automated virus scan, and that that should be sufficient to CYA.

I'm not suggesting that this is always the case. I assume that there are some super top secret places where every byte transmitted over the network is scrutinized by 100 diligent people, because national security depends on it. But I've never personally witnessed such a thing - one reason being that I've never had such a role, so wouldn't know if it was happening.

I believe the codes for the US nuclear silos were all 0000 at one point, until someone found out and complained.

And the US drones operating in Iraq were sending unencrypted video until someone complained about that too.

I'm more likely to be the one complaining - in a careful manner to avoid any individual getting sacked - than the one complained about.

I don't think I would have complained about the missile silos or drones though, as I wouldn't want to take responsibility for the side-effects of changing the code to something that everyone is going to forget in 10 minutes flat.

As Ann Coulter pointed out, and I suspect too - the CIA (public servants) probably operate more like "Paul Bart - Mall Cop" than the super-efficient guys that movies portray them to be. Even at the CIA level - likely everyone just wants to CYA. I don't know for sure - I've never worked for the CIA. I've just met quite a number of people - including respectable professionals. People normally aren't self-confident enough to poke their nose where it probably doesn't belong and run the risk of things backfiring. All for likely no personal benefit. We're social animals designed to want to get along.

I also don't want to give the opposite impression - that it's the Wild West. Auditors exist, and I've been asked tough questions before. And the result is usually a procedural change to prevent a repeat, not a sacking or even a bad review. At the end of the day you need staff to get a job done. You can't sack everyone on a dime. But it's a mix out there. I'm not making hard and fast statements or even able to give percentages. I have only worked for a tiny sliver of employers in the last approx (57-18) years.
Octocontrabass
Member
Member
Posts: 5883
Joined: Mon Mar 25, 2013 7:01 pm

Re: new linux standard

Post by Octocontrabass »

kerravon wrote: Mon Jul 07, 2025 12:35 amWine is already in that situation. The special thing they need is to be bug-free and use only documented features.
No they don't? Sure, Wine works better for programs that are bug-free and use only documented features, but Wine is intended to support any program that Windows supports, including programs that are buggy and programs that use undocumented features.
kerravon wrote: Mon Jul 07, 2025 12:35 amI have basically standardized on Win32 executables. I consider them to be universal already. They run on MSDOS 4.0 + HX. They run on Wine. They run on OS/2 under my PDOS-generic layer. They run on Windows.
That doesn't sound very universal to me. DOS 4 and OS/2 have both been obsolete for decades.
kerravon
Member
Member
Posts: 332
Joined: Fri Nov 17, 2006 5:26 am

Re: new linux standard

Post by kerravon »

Octocontrabass wrote: Mon Jul 07, 2025 12:00 pm
kerravon wrote: Mon Jul 07, 2025 12:35 amWine is already in that situation. The special thing they need is to be bug-free and use only documented features.
No they don't? Sure, Wine works better for programs that are bug-free and use only documented features, but Wine is intended to support any program that Windows supports, including programs that are buggy and programs that use undocumented features.
You can't possibly support "programs that are buggy" - at least not all. It is random how Windows will react to a wild pointer. You can't force Wine to reproduce Windows memory allocation precisely. Even Windows doesn't do that for Windows, and some buggy programs will fail on a Windows upgrade. And with effort you can probably reverse-engineer any undocumented features, but that takes time, and the end result is that stuff like Visual Studio won't run on Wine (or so I've been told).
kerravon wrote: Mon Jul 07, 2025 12:35 amI have basically standardized on Win32 executables. I consider them to be universal already. They run on MSDOS 4.0 + HX. They run on Wine. They run on OS/2 under my PDOS-generic layer. They run on Windows.
That doesn't sound very universal to me. DOS 4 and OS/2 have both been obsolete for decades.
Those were just two environments I listed in that sentence. Previously I listed a lot of others - Wine, Windows etc.

And OS/2 may be a niche environment, but it is being developed and you can actually report bug fixes. A commercially-supported OS where end users can report bugs for free. Basically similar to the mainframe. That may be the best OS available. And the PDOS-generic shim will make (certain) Win32 executables work.

DOS 4 is under active development too (not by Microsoft though - so?). So is HX (again - not by Microsoft - so?). And if they run Win32 executables - what more do you need? That runs more Win32 executables than PDOS/386. It's another great environment. Not only that, but it has the potential to be commercialized to be as good as OS/2, as there are no restrictions on close-sourcing your work to keep it a trade secret.

Win32 may not be strictly universal. But it's a hell of a lot closer to being universal than whatever comes second.
Octocontrabass
Member
Member
Posts: 5883
Joined: Mon Mar 25, 2013 7:01 pm

Re: new linux standard

Post by Octocontrabass »

kerravon wrote: Mon Jul 07, 2025 5:17 pmYou can't possibly support "programs that are buggy" - at least not all.
True, but many bugs are bad assumptions about API behavior. A simple example is relying on an undocumented side-effect, such as an API that "always" calls another function, or an API that "always" returns sorted results. A more complicated example is outright misusing an API, such as pushing the wrong number of parameters on the stack before calling a function. These are, as far as I know, real examples of things some Windows programs will do. (I wish I could remember where I read about that last one. It was quite the mess for whoever had to debug it!)
kerravon wrote: Mon Jul 07, 2025 5:17 pmAnd OS/2 may be a niche environment, but it is being developed and you can actually report bug fixes.
OS/2 is not under development. Are you talking about ArcaOS? Maybe "obsolete" isn't the right word, but it's a 32-bit OS, which is an outdated design.
kerravon wrote: Mon Jul 07, 2025 5:17 pmDOS 4 is under active development too (not by Microsoft though - so?).
DOS is a 16-bit OS, which is even more outdated than OS/2. (And I have no idea where to find the active development you're talking about.)
kerravon wrote: Mon Jul 07, 2025 5:17 pmSo is HX (again - not by Microsoft - so?).
What the heck is HX?
kerravon wrote: Mon Jul 07, 2025 5:17 pmNot only that, but it has the potential to be commercialized to be as good as OS/2, as there are no restrictions on close-sourcing your work to keep it a trade secret.
Which OS has restrictions on closed-source software?
Post Reply