Is anything in IT NOT shit?
-
@Captain said:
Vim's not shit. It's way better than notepad.
Everything's better than notepad.
Have you used ed?
-
I was only aiming for my niece.
, which is almost more terrible considering the replied-to comment...
-
I'd say you have my sympathies but I'm in the same boat - only in PHP.
-
This post is deleted!
-
There's IntelliSense in the IDE, which is pretty damn good IMO.
And there's an add-in for Visual Studio.
-
almost more terrible considering the replied-to comment..
Dammit! Over shot it again!
-
@Zacrath said:
@Captain said:
Vim's not shit. It's way better than notepad.
Everything's better than notepad.
Have you used ed?
Still better than notepad. notepad's features, unlike other editors, cause problems instead of solving them.
Filed under: Stubbornly sticking to my guns here
-
Academic research is anything but fun. It is too far on the hard scale.
Maybe a-team wasn't a good analogy, it would have to be more like some common mercenaries that die in the first half of the movie.
-
But Notepad was never intended to be a full fledged text editor, it's just a demo of the multiline textbox control. The same goes for WordPad (formerly Write) which is just there to show off the RTF editor.
You can even find the source code for WordPad somewhere on Microsoft's website.
-
IIRC, I once slapped a multiline textbox control in a form in delphi 2, and the result was actually better than notepad, though I don't recall the key additional feature. Probably loading files with newline-only line endings.
-
I was only aiming for my niece.
I ... have no response for that, but it's certainly not the direction I intended this to go.
-
And don't forget, 50% of all shit is IT or IT related. Or just IT.
-
source code for WordPad
Oh wow, it does exist!
It is an example of a full-fledged word processing application written using MFC.
Well it's a good thing the page didn't call it a "full flexed text editor"...
-
-
certainly not the direction I intended
It wasn't the direction I was aiming for either, but here we are ...
-
a good project lead that finished on time
In my experience, they always work late
And if all of this fails, the biggest problem when switching IT jobs is matching your current salary
I always manage to get at least a 5K increase when I move jobs. I guess I just interview well
-
Until you reach your zone salary limit. This happened for me too when I started, but now that is harder. For example, around here the avg salary of IT workers is around 30k-36k euros. Anything above that is difficult to score.
-
It was so well planned, and the customer was so well invested in the project, that it was a total anti-climax that everything went exactly to plan!
Note the parameters of a successful project:
- Clear, precise requirements
- No business people involved, purely technical operation
- Plenty of time
- Competent, permanent team involved (no contractors, revolving doors, etc.)
-
Until you reach your zone salary limit. This happened for me too when I started, but now that is harder. For example, around here the avg salary of IT workers is around 30k-36k euros. Anything above that is difficult to score
I probably haven't got the peak yet then. I know I could get more working in London but no amount of money is worth heading into that hellhole on a regular basis.
-
I wouldn't recommend it. I did it for 18 months, never again.
-
In my experience, they always work late
I did have worked with a PM who can actually lead the project to finish on time on budget. The project under his lead blocks any unreasonable requests. All team members in the team are experienced developers contribute to the success in progress control too.Unfortunately because of the way he blocks requests, he is determined by the upper managements to be "not helpful and in fact prevents the company to accompanish it's mission" so he is sent away.
That's why we can't have nice things.
-
-
Unfortunately because of the way he blocks requests, he is determined by the upper managements to be "not helpful and in fact prevents the company to accompanish it's mission" so he is sent away.
As always, The Real WTF (TM) is upper management.
-
There's an open source program that I think is good, nice, intuitive, and works well: GParted. Probably the one program on Linux that managed to exceed my expectations.
I like the dialog when you apply the operations. You can expand each part and see exactly what it does.
...and yet, it still managed to (almost) trash my disk once, since one of the CLI programs it calls underneath changed its interface to add a confirmation prompt, and GParted got confused and just gave up. That's what the Unix philosophy gets you.
-
...and yet, it still managed to (almost) trash my disk once, since one of the CLI programs it calls underneath changed its interface to add a confirmation prompt, and GParted got confused and just gave up. That's what the Unix philosophy gets you.
As in the 'do one thing and do it well' philosophy? In the case of a front-end UI like GParted to a set of commandline tools, that philosphy means said UI needs to do one thing well: correctly interface between the user and the underlying commandline tools.
Obviously, the failure here is on the part of the GParted developers, who failed to update the call to the underlying CLI program and/or on the part of the package maintainers, who didn't limit compatibility to the correct subrange of the CLI utility in question.
The 'do one thing and do it well' philosophy is a pretty damn important aspect of SOLID and one of the things *-nix actually got right. (*-nix and its eco-system of software development got a metric buttload of stuff wrong to counterbalance that though...)
-
The bad philosophy is "CLI can be both a human interface and an API for other programs".
Because then it means you can never change anything from a program ever, or you break all existing programs that use it. And "update them all" is not a solution.
-
a front-end UI like GParted to a set of commandline tools,
Ding ding ding!
Congratulations, you have already described a huge anti-pattern that guarantees both the GUI and CLI for this particular functionality will be shit forever!
Using the CLI as an API means you've now made it fucking impossible to ever improve the CLI. Because making any change to it will break an unknown number of programs that rely on its behavior.
Why the fuck doesn't this application have an Application Programming Interface that both the CLI and GUI use? Oh right: because Linux developers are really shitty at developing software. I almost forgot for a nanosecond.
-
Not that Linux has a monopoly on that. Exchange 2010 GUI/PowerShell says hi. (2013 is somewhat better, but still frustrating).
-
-
Academic research is anything but fun. It is too far on the hard scale.
Maybe you're just underrating your ability? Also, someone has to provide IT and software engineering support to the researchers…
-
Because then it means you can never change anything from a program ever, or you break all existing programs that use it. And "update them all" is not a solution.
Well, even if UI didn't interact with the program over CLI the problem still stands. You may not break the UI layer but what about all the shell scripts and what-have-you that rely on the CLI working a certain way?
Let's face it, you can't change any interface ever without breaking something. Even changing the GUI may result in some confused grandma being unable to use the internets.
-
Well, even if UI didn't interact with the program over CLI the problem still stands. You may not break the UI layer but what about all the shell scripts and what-have-you that rely on the CLI working a certain way?
That's why some CLI programs have a switch that enables a stable command-line API for shell scripts. IMO that's the only sane way of dealing with that particular problem. Still not an excuse for not writing a library with a proper API, though.
-
Let's face it, you can't change any interface ever without breaking something.
That's why it is a really good idea to never rely on things failing in a particular way; the space of failures is where there is space for expansion of functionality in the future.
-
Congratulations, you have already described a huge anti-pattern that guarantees both the GUI and CLI for this particular functionality will be shit forever!
Yep.
Because"CLI can be both a human interface and an API for other programs"
is indeed one of those things I mentioned that came from the metric buttload of stuff that was wrong with *-nix.
The GParted developers screwed up by depending directly on the commandline tools, rather than some abstracted form of API, sure. Ofcourse, such an API doesn't exist, because this is *-nix with its warped sense of centralizing the commandline as the end-all-be-all of thing. So really; what was their choice?
-
So really; what was their choice?
Use an OS that doesn't suck ass?
Write software that makes the OS suck less ass instead of just giving up and increasing the ass-suckage?
Seppuku?
All better choices.
-
Use an OS that doesn't suck ass?
This doesn't jive with your advice to upgrade to Windows 10.
-
What advice?
-
What advice?
Breaking news: Blakeyrat flip-fops; slams Windows 10, gushes about Linux.
-
Use an OS that doesn't suck ass?
If your goal is to write a set of partitioning tools for a particular OS, then writing it to actually run on that OS is something that's kind of a given...
