Active Topics
-
Sailfish OS for the Motorola Moto G7 Power (XT1955-5) - (ocean) (9)
to SailfishOS by biketool - 18 hrs ago -
[Announce] coderus unrestricted system-ui (297)
to MeeGo / Harmattan by nieldk - 6 days, 23 hrs ago - more...
 |
|
2009-10-11
, 12:11
|
|
Posts: 3,159 |
Thanked: 2,023 times |
Joined on Feb 2008
@ Finland
|
#22
|
Originally Posted by AVee
If one is determined to install unstable software, he will do that no matter what disclaimers there are. I bet that even when disclaimer is negative, there will be enough testers (at least guys who now own n8x0s and are part of community and somewhat familiar about maemo).
Personally, I hate those kind of warnings. They are addressing the issue from the wrong (=negative) angle.
Suppose I'm not a developer, I'm not an experienced tester and I really don't know what I'm doing. How is that ever going to change if I follow warnings like this? The only way to learn what is going on is to ignore warning like this and try it anyway. The only way to become an experienced tester is to test software which actually has bugs, etc.
The warning should really be phrased positive, e.g. "Feel free to play with this software, but be aware that..." And add the 'Restore factory image' instructions in there, doing so will scare more people away then any warning will.
Positive warning sounds like sleazy car salesman that sells you $2000 car for $5000 and gives wrong contact info so warranty is void before your taillights disappear. ("this is just formality to sign here, the text doesn't matter, not worth reading it through. Just sign! ... I promise there will be no problems.")
It totally is the wrong way to get testers. Luring innocent fools to get their shiny new devices to a point that a reflash is needed is not the publicity maemo community needs...
| The Following 2 Users Say Thank You to ossipena For This Useful Post: | ||
|
|
2009-10-11
, 12:30
|
|
Posts: 8 |
Thanked: 4 times |
Joined on Oct 2009
|
#23
|
I'd like to add:
You've only one chance to get users convinced that it doesn't matter that their phone runs Linux.
I have experience with a lot of Linux based devices in the past. First thinking: wow you can write your own software blabla.. But really, there was no one that was really open, always changing APIs, not stable... Bottom line: I'm not an average customer, but even I sold all these devices almost directly because they were totally un-customer-friendly.
This is not what an average customer want. Customers ARE NO BETA TESTERS, they pay hard money.
If we want to make (money with) nice applications in the future, based on Linux, then we need to remember this I think.
Remember: we're Maemo, we're the (Linux) future :-)
If we do it well we can beat Microsoft and all propriatary OSes in the market, if we fail then..... Then we fail (forever?).
Joep
You've only one chance to get users convinced that it doesn't matter that their phone runs Linux.
I have experience with a lot of Linux based devices in the past. First thinking: wow you can write your own software blabla.. But really, there was no one that was really open, always changing APIs, not stable... Bottom line: I'm not an average customer, but even I sold all these devices almost directly because they were totally un-customer-friendly.
This is not what an average customer want. Customers ARE NO BETA TESTERS, they pay hard money.
If we want to make (money with) nice applications in the future, based on Linux, then we need to remember this I think.
Remember: we're Maemo, we're the (Linux) future :-)
If we do it well we can beat Microsoft and all propriatary OSes in the market, if we fail then..... Then we fail (forever?).
Joep
Originally Posted by ossipena
If one is determined to install unstable software, he will do that no matter what disclaimers there are. I bet that even when disclaimer is negative, there will be enough testers (at least guys who now own n8x0s and are part of community and somewhat familiar about maemo).
Positive warning sounds like sleazy car salesman that sells you $2000 car for $5000 and gives wrong contact info so warranty is void before your taillights disappear. ("this is just formality to sign here, the text doesn't matter, not worth reading it through. Just sign! ... I promise there will be no problems.")
It totally is the wrong way to get testers. Luring innocent fools to get their shiny new devices to a point that a reflash is needed is not the publicity maemo community needs...
| The Following User Says Thank You to duckyduck For This Useful Post: | ||
 |
|
2009-10-12
, 08:51
|
|
Posts: 3,105 |
Thanked: 11,088 times |
Joined on Jul 2007
@ Mountain View (CA, USA)
|
#24
|
Originally Posted by duckyduck
Because in a Debian GNU/Linux system the root partition is not a "wrong location" at all, and if we would put a tiger policy probably many developers would be quite upset.
The question I have is: why is the package manager not smart enough to refuse to install applications in the wrong location? (with a special developers mode that you need to enable by shell or something) Or simply correcting the install path.. It would force the developer to think about and install in the correct location.
This is why we decided not to enforce but to recommend developers to use /opt and /home/MyDocs (in the eMMC) as much as possible.
The idea of having a common warning header in the threads discussing an unstable application is very good. Most of the "innocent users" jumping to unstable software will do so after reading a Talk thread or whatever someone wrote elsewhere after seeing a Talk thread.
I agree with the tone: it's much better to have something short and positive encouraging testing & feedback BUT acknowledging the risks. A "If you break something don't blame us since you've been just warned"
or something should be enough.
| The Following User Says Thank You to qgil For This Useful Post: | ||
|
|
2009-10-12
, 09:00
|
|
Posts: 883 |
Thanked: 980 times |
Joined on Jul 2007
@ Bern, Switzerland
|
#25
|
Perhaps it would make sense to make the App Manager more "aware" of certain repositories. Extras-devel and -test shouldn't have .install files in my opinion, so that they would neeed to be added manually. And installing from these could also show an even more impressive disclaimer.
| The Following User Says Thank You to twaelti For This Useful Post: | ||
|
|
2009-10-12
, 09:31
|
|
Posts: 337 |
Thanked: 160 times |
Joined on Aug 2009
@ München, DE
|
#26
|
Originally Posted by pycage
Although this strays a bit away from the original meaning, I think this should be written as:
In German:
Warnung
Diese beiden Quellen sollten nur von Entwicklern oder erfahrenen Benutzern eingetragen werden.
Software aus diesen beiden Quellen sollte nur von Entwicklern oder erfahrenen Benutzern installiert werden.
(Software from these two repositories should only be installed by developers and "seasoned" users).
Doesn't sound as awkward to me.
Sofern nicht anders angegeben, kann nicht ausgeschlossen werden, dass alle in diesem Thread erwähnten Programme die Funktionsweise Ihres Gerätes ernsthaft beeinträchtigen könnten.
The part before the first "," probably isn't needed - and "alle in diesem Thread erwähnten" (all software mentioned in this thread) is a factual mistake - how should anyone know *which* thread *where* is meant by that?
My 2 Eurocents ...
| The Following User Says Thank You to range For This Useful Post: | ||
 |
|
2009-10-12
, 10:15
|
|
Posts: 2,121 |
Thanked: 1,540 times |
Joined on Mar 2008
@ Oxford, UK
|
#27
|
Originally Posted by twaelti
I completely agree with this in principle. "Real" developers and testers will have no worries about manually typing in such repositories. On the other hand, although I think we should take down the .install links from, for example, http://repository.maemo.org/, I think it would be over the top to ban them completely from existing on other websites, or to stop h-a-m from processing them.
Extras-devel and -test shouldn't have .install files in my opinion, so that they would neeed to be added manually.
 |
|
2009-10-12
, 10:51
|
|
Posts: 3,404 |
Thanked: 4,474 times |
Joined on Oct 2005
@ Germany
|
#28
|
Aren't the .install files a relic of the old days with lots of different repositories anyway?
The idea behind an .install file (to install application from webbrowser) is good but it should not modify the list of repositories anymore. The system should only accept .install files for repositories that the user has already configured, IMHO.
The idea behind an .install file (to install application from webbrowser) is good but it should not modify the list of repositories anymore. The system should only accept .install files for repositories that the user has already configured, IMHO.
__________________
Tidings - RSS and Podcast aggregator for Jolla - https://github.com/pycage/tidings
Cargo Dock - file/cloud manager for Jolla - https://github.com/pycage/cargodock
Tidings - RSS and Podcast aggregator for Jolla - https://github.com/pycage/tidings
Cargo Dock - file/cloud manager for Jolla - https://github.com/pycage/cargodock
|
|
2009-10-12
, 11:00
|
|
Posts: 3,841 |
Thanked: 1,079 times |
Joined on Nov 2006
|
#29
|
I think that's too strict. It should IMO allow it, but with a pop-up about the fact that it's going to add a repo to the repo list, yes/no?
__________________
N800/OS2007|N900/Maemo5
-- Metalayer-crawler delenda est.
-- Current state: Fed up with everything MeeGo.
N800/OS2007|N900/Maemo5
-- Metalayer-crawler delenda est.
-- Current state: Fed up with everything MeeGo.
| The Following 3 Users Say Thank You to TA-t3 For This Useful Post: | ||
 |
|
2009-10-12
, 11:22
|
|
Posts: 2,869 |
Thanked: 1,784 times |
Joined on Feb 2007
@ Po' Bo'. PA
|
#30
|
Originally Posted by pycage
Aren't the .install files a relic of the old days with lots of different repositories anyway?
The idea behind an .install file (to install application from webbrowser) is good but it should not modify the list of repositories anymore. The system should only accept .install files for repositories that the user has already configured, IMHO.
Agreed... a warning of sorts is triggered on the device when a new repository or catalog needs to be installed. Can this dialogue be changed or can selected dialog be produced if triggered by the catalogs title or source?
No matter what the warning, imho something should be posted here, in a highly visable place, soon... it can always be edited later.
The real danger is if a popular blog posts a link, or copies one of these install files to their site with out regard.
Last edited by YoDude; 2009-10-12 at 11:26.
I realize that's a separate issue, but it's also part of the user experience that isn't going to impress the new crowd.