Page 23 of 31 « First 132122 23 2425 Last »
Reply
Thread Tools
ewan's Avatar
ewan is offline ewan
2010-02-19 , 12:22
Posts: 445 | Thanked: 572 times | Joined on Oct 2009 @ Oxford
#221
Originally Posted by twaelti View Post
I'm STILL astonished about the "we-are-FOSS-and-so-much-better-than-the-rest-of-this-greedy-world" behaviour shown here. This thread reminds me of a communist mob. Stop dreaming and being naive. Very few of the available FOSS software came into existence for "free".
I have absolutely no idea what you're responding to with this sort of thing. This has never been a free vs paid or Free vs proprietary dispute. This is entirely about there being a process designed to ensure a degree of QA of software released to the public repository, and that process being circumvented.

That's all.
 

The Following 7 Users Say Thank You to ewan For This Useful Post:
Freemantle is offline Freemantle
2010-02-19 , 13:08
Posts: 94 | Thanked: 44 times | Joined on Feb 2010
#222
vote for thread lock

with new threads to carry on "FOSS vs the world" and QA improvement discussions.
 

The Following User Says Thank You to Freemantle For This Useful Post:
tomster is offline tomster
2010-02-19 , 13:27
Posts: 71 | Thanked: 33 times | Joined on Dec 2009 @ Munich, Germany
#223
Originally Posted by ewan View Post
This is entirely about there being a process designed to ensure a degree of QA of software released to the public repository, and that process being circumvented.
Hasn't the mentioned vetting process proven to be flawful to some extent?

By outcome I think there's not much difference in asking 10 random voters to vote for your repo-file (as suggested some 90'ish postings back) or to set-up 10 fake users doing exactly the same. Still, by outcome that'd be.

A hurdle not being a hurdle is, well.. not a hurdle...

As much as I do remember there would still have to be 3 votes from a "testers group" anyway (whatever or whoever that would be) to get the package actually promoted. Still there's not near 70% assurance that the package is really ready for a broadend number of users.

BUT even your (excuse the povocative word) hallowed vetting process can't seriously save our phones from harm. Being a huge supporter of the open-source idea and especially unlimited control over my device, I have to admit that as of now merely everybody could just gain the same powers...

Compared to e.g. the iPhone with its "limited" API access and sort of "containered" app-handling we're surely facing bigger security issues with the N900/maemo. This might be the time to think about a container-like API allowing only "unharmful" operations by future apps. I know this might offend a considerate amount of you out there, but still...

Not only to ease QA processes, by not having to look into every detail, but much more to really assure that apps can't break into your device. Also think in a couple of months time, when maemo, sorry meego, should become a fool-proof end-user OS.
Last edited by tomster; 2010-02-19 at 13:29.
 

The Following User Says Thank You to tomster For This Useful Post:
benny1967's Avatar
benny1967 is offline benny1967
2010-02-19 , 13:40
Posts: 3,790 | Thanked: 5,718 times | Joined on Mar 2006 @ Vienna, Austria
#224
Originally Posted by tomster View Post
Pardon my French: Congratulations for beating the first real 3rd party game provider out the house with a toilet brush.
It doesn't matter at all if he/she develops games or a word processing application or a browser. It also doesn't matter at all if the application is commercial, non-commercial, free, proprietary, 3rd party, 5th party or no party.

What does matter is it is the first known (!important!) case of a developer not playing by the rules. I wonder how the people feel whose applications decay in -testing because nobody is interested in testing them... And they still wait and hope. And now somebody comes along and skips the whole process. If we don't beat him out the house, others will follow his example... and tomorrow we'll have no Q&A process any more.

The whole thing is healthy because it shows where the limits are. It shows that there are consequences.
 

The Following 12 Users Say Thank You to benny1967 For This Useful Post:
ewan's Avatar
ewan is offline ewan
2010-02-19 , 13:55
Posts: 445 | Thanked: 572 times | Joined on Oct 2009 @ Oxford
#225
Originally Posted by tomster View Post
Hasn't the mentioned vetting process proven to be flawful to some extent?

By outcome I think there's not much difference in asking 10 random voters to vote for your repo-file (as suggested some 90'ish postings back) or to set-up 10 fake users doing exactly the same. Still, by outcome that'd be.
I don't think there's a flaw in the process. I quite understand that it's easy to be incredulous when faced with the idea that requiring ten votes from random people is any sort of worthwhile quality check, but the simple fact is that it demonstrably works. It's worked for Maemo up until now, successfully filtering a lot of interesting, but distinctly not-ready-for-the-masses software from getting from extras-devel and -testing into extras. Similar processes work for other community maintained software repositories, and there's a lot of them around now.

If you get ten votes from random interested people you do actually have a reasonable assurance that there aren't glaring howlers in the code. The case of getting ten votes from the original author of the software essentially means that no testing has taken place at all, and that's clearly a very different situation.

It is (obviously) possible to deliberately bypass the checks, but I'm not sure there's any way top prevent that a priori without also excluding would-be testers from participating. It is possible to watch for and fix abuses after they've happened, and that's exactly what happened here. I think it is worth having a more formal process for dealing with this situation, and that's part of what we're in the process of discussing.
 
Texrat's Avatar
Texrat is offline Texrat
2010-02-19 , 15:33
Posts: 11,700 | Thanked: 10,045 times | Joined on Jun 2006 @ North Texas, USA
#226
Originally Posted by benny1967;536543What does matter is it is the first [I
known[/I] (!important!) case of a developer not playing by the rules. I wonder how the people feel whose applications decay in -testing because nobody is interested in testing them... And they still wait and hope. And now somebody comes along and skips the whole process. If we don't beat him out the house, others will follow his example... and tomorrow we'll have no Q&A process any more.

The whole thing is healthy because it shows where the limits are. It shows that there are consequences.
I agree with the reason behind the outrage... but wasn't the cure worse than the disease?
__________________
Nokia Developer Champion
Different <> Wrong | Listen - Judgment = Progress | People + Trust = Success
My personal site: http://texrat.net
 

The Following 3 Users Say Thank You to Texrat For This Useful Post:
benny1967's Avatar
benny1967 is offline benny1967
2010-02-19 , 15:42
Posts: 3,790 | Thanked: 5,718 times | Joined on Mar 2006 @ Vienna, Austria
#227
Originally Posted by Texrat View Post
I agree with the reason behind the outrage... but wasn't the cure worse than the disease?
I don't think so, no.

Because I don't compare two applications gone to one developer betraying the community and skipping the Q&A process.

I compare two applications gone to all developers betraying the community and skipping the Q&A process int he future.

What reason would you have to wait for people to test and vote for your application if you can do it yourself?
 
Texrat's Avatar
Texrat is offline Texrat
2010-02-19 , 15:45
Posts: 11,700 | Thanked: 10,045 times | Joined on Jun 2006 @ North Texas, USA
#228
Well, I'm not going to help build any straw men... I just think some of the response has been disproportionate to the offense.

I'd rather first try to correct bad behavior, then burn a repeat offender at the stake.

On a side but related note: I keep seeing the word "open" thrown about here rather carelessly. Open in this regard should certainly not mean "open to abuse". It should mainly mean "open to inspection". A highly-visible process with reasonable safeguards.
__________________
Nokia Developer Champion
Different <> Wrong | Listen - Judgment = Progress | People + Trust = Success
My personal site: http://texrat.net
Last edited by CrashandDie; 2010-02-19 at 15:49. Reason: double post
 

The Following 9 Users Say Thank You to Texrat For This Useful Post:
johnel is offline johnel
2010-02-19 , 15:55
Posts: 999 | Thanked: 1,117 times | Joined on Dec 2009 @ earth?
#229
I'd rather first try to correct bad behavior, then burn a repeat offender at the stake.
This is great news, we can toast marshmellows and sing songs round the campfire.

or

Maybe we can build a wickerman instead?
We can then dance like pagans but we need to decide who will play Christopher Lee first.

We can still toast marshmellows.
__________________
I like cake.
 
ewan's Avatar
ewan is offline ewan
2010-02-19 , 16:15
Posts: 445 | Thanked: 572 times | Joined on Oct 2009 @ Oxford
#230
Originally Posted by Texrat View Post
Well, I'm not going to help build any straw men... I just think some of the response has been disproportionate to the offense.

I'd rather first try to correct bad behavior, then burn a repeat offender at the stake.
That sounds fair, but there's still a risk of things going this way - if the developer reacts badly to being corrected and goes off in a strop, that's up to them. I'm not sure that there's any procedure that we could have that could keep that from happening.

On a side but related note: I keep seeing the word "open" thrown about here rather carelessly. Open in this regard should certainly not mean "open to abuse". It should mainly mean "open to inspection". A highly-visible process with reasonable safeguards.
I completely agree, but any process that's going to be openly visible is going to have the potential for people who are upset expressing their feelings. I think the transparency is more important that avoiding the appearance of unpleasantness in public, but I imagine not everyone will agree with that.

I do think that the substance of the response in this case (i.e. apps dropped, upload privileges revoked, explanation required) is the right response. It should just probably be written down somewhere.
 

The Following 3 Users Say Thank You to ewan For This Useful Post:
Reply
Page 23 of 31 « First 132122 23 2425 Last »

« Previous Thread | Next Thread »
Thread Tools

 
Forum Jump


All times are GMT. The time now is 04:43.

Contact Us - Home - Archive - Top