|
If I use Open Source, it is to understand how to solve that specific problem. The work I put in to understand it usually leads me to rewriting the code to suit my specific needs in a better way, and I end up with something that does not resemble the original open source code.
I would never trust code that I do not understand the workings of, just because it is labeled 'open source'. OS code that I have not studied and fully understood myself, is functionally equivalent to closed source code, similar to a compiled library.
Closed code is only trustworthy if you know that the source is trustworthy. Open Source that you have not studied and fully understood yourself is trustworthy only if the source is trustworthy.
Common OS repositories are not necessarily trustworthy.
|
|
|
|
|
I have open source repos, use bitbucket, contribute, fix bugs AND fork for my own needs...
|
|
|
|
|
In my experience, there are three types of contributors:
- Developers at the start of their career - trying to build a portfolio
- Developers at the end of their career - keeping their hand in
- Developers who are paid by their employer to add features, e.g. drivers under Linux or academic projects
Very few if any people are willing to spend hours a week maintaining a product for no pay and no prospects of pay.
Freedom is the freedom to say that two plus two make four. If that is granted, all else follows.
-- 6079 Smith W.
|
|
|
|
|
Which results in
1) No competence;
2) No interest;
3) The only usable stuff.
GCS/GE d--(d) s-/+ a C+++ U+++ P-- L+@ E-- W+++ N+ o+ K- w+++ O? M-- V? PS+ PE Y+ PGP t+ 5? X R+++ tv-- b+(+++) DI+++ D++ G e++ h--- r+++ y+++* Weapons extension: ma- k++ F+2 X
|
|
|
|
|
Exactly!
Freedom is the freedom to say that two plus two make four. If that is granted, all else follows.
-- 6079 Smith W.
|
|
|
|
|
... via emails and forums.
|
|
|
|
|
I don't have time to contribute at home. At work, not only am I busy, but often cannot contribute due to restrictions on what I do. But, sometimes, a developer provides a link for donations, and I contribute a little.
|
|
|
|
|
Ope ... there's no lawyer on the planet that's going to tell me that I missued someone elses code; so this survey is complete bollox
|
|
|
|
|
You mean you feel you'll never be sued for using someone else's code? Because no case would hold water, or you don't use OS code?
cheers
Chris Maunder
|
|
|
|
|
like like like like (head explodes at this point (see "Scanners" (the movie)) ...
I live in a city where there isn't a person walking down the sidewalk that isn't looking through his cellphone at his feet lining up on my corpus about to headlong buttheads with me, the only citizen in that city without a cellphone ... paying attention. And I also have to listen to this drone of "like" concentrating on the porches of my ears at that same time so I turn to thinking about how there's no law that says I can't misuse Open Source code.
And I've come to the conclusion that this is mainly due to the fact that code of any kind can't really be misused. And not because the word "code" hasn't been defined.
|
|
|
|
|
Like the first response said, it can help disencumber “client owned” branches.
I have run into that issue multiple times on multiple open source projects. Since I billed the client for the fixes needed, I needed their permission to release the changes. I never received a release to publish.
In hindsight, I should NOT have billed for those changes. This was before the git model allowed for easy public forking. You had to email changes to maintainers.
One defect had a simple, missing overrun check. I checked back on the code a few years later and someone had contributed a totally awful implementation to try and fix the same problem. I doubt the contributor understood the root cause.
|
|
|
|
|
Interesting.
So if you don't bill then your fix is on your own time and there's no red tape.
I'm surprised a client allowed you to use Open Source code., at least not without a formal release from the author or a release from their legal dept about including third party code that doesn't have a picture of the person they can sue pinned next to it.
Lawyers will always go for the biggest fish, and some poor developer giving away his code is far less a realistic target for damages than the big corp that used said developer's code, so the legal departments always are always trying to make sure they are not the juiciest targets.
It's like when you're being chased by a bear. As long as your second slowest or better you're OK.
cheers
Chris Maunder
|
|
|
|
|
You have to read the license attached to the project. I have passed on many open source uses due to requirements to release the other parts of the system.
|
|
|
|
|
I chose the download and don't contribute answer and I thought to myself "does this make me feel bad?" The answer is truthful. I'd venture we all download and use "open source projects". Heck, .net is open source as well as many standard libraries in most languages.
So, I download and use them. Yep. Beyond those, I also think that for many of us, it's not worth the time to venture into OS projects to fulfill a need we have. In the past ten or so years, I've found maybe two OS things that were worthwhile for projects I was working on. And even then, by the time I amended and extended them to make them usable, I wasn't sure that writing from scratch wouldn't have been better. I didn't contribute back as my use cases are narrow, boring, and I really don't have the time to get into the often ridiculous, self-righteous world of git or OS sites. Stack Exchange is an even worse rabbit-hole, but that would be a subject in itself.
|
|
|
|
|
I take the time to write out issues I encounter and I patch them.
If it's for a client's project, I will branch from whatever repo is available and patch whatever I need patched, so I can use the work in future projects, without being tied to a specific client.
I often meet students who approach the idea of Open Source as a principle, which is a flawed way to approach to subject.
Open Source is a mechanism to decouple valuable bits of code from specific products or clients.
And it's a way to gather value amongst different use-cases.
|
|
|
|
|
Isn't writing an article along with code, on this CP site, considered as Open Source? I feel the answer is Yes.
|
|
|
|
|
I would agree
However, articles under CPOL are not considered "Open Source" in that they don't provide full Freedom. They are free (as in no cost), they provide all code, they can be used and redistributed, but we include a clause that says the code can't be used for illegal or immoral purposes.
We do this because at the time we drafted that license there had been cases where a developer had given away code, that code had been used for illegal purposes, and that developer was then sued. (It actually reminds me of a recent case where a cyclist was struck by a car that ran a red light and the cyclist, after getting out of hospital, was handed a bill from the car's insurance company).
So we made the decision that providing our authors with something they can hold up in court and say "I didn't condone this!", while enabling anyone, anywhere to use their code for whatever legal and moral reason they want, is worthwhile. Those who want to use the code for illegal and/or immoral reasons probably aren't going to care about the license.
So at a practical level anyone can use the code on CodeProject. From a legal perspective there's a small protection for the author. The Open Source community (as in the FOSS community) would rather say that the communities right to do what they want outweighs the need to allow an individual's legal protection.
So long story short: we built CodeProject to allow safe sharing of code and knowledge. Code on CodeProject is Open (it can be viewed), it contains Source. It costs no money. It just doesn't give the user the right to use the code in a manner that may legally impact the author.
cheers
Chris Maunder
|
|
|
|
|
Thanks, Chris, for the detailed clarification.
|
|
|
|
|
Amarnath S wrote: Thanks, Chris, for the detailed clarification angry old man rant
FTFY.
cheers
Chris Maunder
|
|
|
|
|
Maybe checkboxes would have been better.
|
|
|
|
|
Clearly.
I would have checked four boxes.
Truth,
James
|
|
|
|
|
... now I don't even use Open Source apps, let alone download the code.
There is not real control, nobody works on the "boring stuff", just the flashy bits they want to use.
That's why I stopped using Open Office, and pay MS £75 a year for the privilege of not owning their software instead. It's just more hassle to rely on OS apps than to pay for the "real thing".
Yes, I know - I could contribute and do the boring stuff. If I had time to learn the code. If the others that contribute didn't have an attitude of "the bit I wanted works, sod you" I might try to find the time. But ... that's as likely as a student in QA doing his own homework ...
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
"Common sense is so rare these days, it should be classified as a super power" - Random T-shirt
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
I would agree with this in general, but there are open source projects for which there are no closed dev corollaries.
An example would be the graphics library I built for IoT, which I did to fill a commercial need that other graphics libraries were not filling.
Now I could have developed that on the clock and held it back, but I've learned that's a mixed blessing, particularly for a contractor like myself who benefits from past work in future projects with future clients. But also I like to contribute to the community, and I felt like I could bring some value.
The problems with open source you list are accurate, and part of that is the low bar for entry. All you have to do to make an open source project is check something into github, give it a license template and promote it somewhere.
That's both good and bad though. If it weren't for the ease of getting an open source project off the ground, a lot of vital projects simply would not exist.
*edits - i'm off my feed today*
To err is human. Fortune favors the monsters.
|
|
|
|
|
Once I encountered an issue with basic or digest authentication (don't remember) and accented chars (basically, the problem was that only ASCII chars were allowed).
This happened several years ago.
I fixed the problem, but some "boss" said that it was not standard behavior, and may break existing sites.
OK, it might have made sense, although I did not understand what could have been broken. Anyway, I added a flag in Firefox options, but someone said that they don't like flags, because it makes the product more complicated (too many flags, something like that).
I had spent several weeks on this problem, first to download the whole source code, to compile it, to understand where the problem could be fixed, and to fix it. And it was finally a total waste of time.
In the end I deleted everything, and decided that I would never try to contribute again (at least to Firefox): I have a *paid* job, why would I give time and work on something which has been finally completely useless.
May be this issue has been fixed... I don't care now.
|
|
|
|
|
A large percentage of OS code is non functional, incomplete or garbage.
A large percentage of OS code is useless.
The percentage of non garbage non useless code is small to begin with... then IP and licensing issues arise. If those are manageable, support, liability and safety certification issues arise.
So thanks but no, thanks.
GCS/GE d--(d) s-/+ a C+++ U+++ P-- L+@ E-- W+++ N+ o+ K- w+++ O? M-- V? PS+ PE Y+ PGP t+ 5? X R+++ tv-- b+(+++) DI+++ D++ G e++ h--- r+++ y+++* Weapons extension: ma- k++ F+2 X
|
|
|
|
|