Sunbird 0.3a1

classic Classic list List threaded Threaded
14 messages Options
Reply | Threaded
Open this post in threaded view
|

Sunbird 0.3a1

Jim Scott
How stable is the new Sunbird alpha.
I would like to use it but all my 'stuff' is on 0.2 and I don't really want
to lose it.
--
Jim
         Tyneside UK
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Christopher Jahn
Jim Scott <[hidden email]> wrote in
news:6hribd17nhtx$.[hidden email]:

> How stable is the new Sunbird alpha.
> I would like to use it but all my 'stuff' is on 0.2 and I don't really
> want to lose it.

Stick with 0.2 for now.

--
}:-)       Christopher Jahn
{:-(       http://home.comcast.net/~xjahn/Main.html

The problem with troubleshooting is that real trouble shoots back.
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Jim Scott
On 3 Dec 2005 14:41:15 GMT, Christopher Jahn wrote:

> Jim Scott <[hidden email]> wrote in
> news:6hribd17nhtx$.[hidden email]:
>
>> How stable is the new Sunbird alpha.
>> I would like to use it but all my 'stuff' is on 0.2 and I don't really
>> want to lose it.
>
> Stick with 0.2 for now.

Thank you.
--
Jim
         Tyneside UK
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Simon Paquet
In reply to this post by Christopher Jahn
Christopher Jahn wrote on 03.12.2005:

>> How stable is the new Sunbird alpha.
>> I would like to use it but all my 'stuff' is on 0.2 and I don't really
>> want to lose it.
>
> Stick with 0.2 for now.

Any reasons for that suggestion?

IMO 0.3a1 is definitely superior to 0.2. It is faster at least as stable
and has a lot of annoying bugs fixed that plagues 0.2. It has some new
bugs, sure, but no real showstopper IMO.

--
Simon Paquet
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Rod Whiteley
Simon Paquet wrote:
> ...IMO 0.3a1 is...at least as stable...

It's true that serious problems seem to have been rare.  For some
examples, see:

http://forums.mozillazine.org/viewtopic.php?t=346279

However, with 0.2 I was able to fix some problems using just a text
editor and the RFC.  I have looked in 0.3a1's data file, and I reckon
any problems would be much harder to fix.  So it seems to me that 0.3a1
needs to be much more stable than 0.2 to give the same level of
confidence.  I don't think we can be certain of this yet.

Using the alpha, but keeping ICS backups might be a good compromise.

--
Rod Whiteley
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Ilja Sekler
Rod Whiteley wrote:
> Simon Paquet wrote:
>> ...IMO 0.3a1 is...at least as stable...

Only one crash during the last month - using nightly builds (0.3a1+).

> It's true that serious problems seem to have been rare.  For some
> examples, see:
>
> http://forums.mozillazine.org/viewtopic.php?t=346279

If it were really tasks and not events that got lost, it might be
https://bugzilla.mozilla.org/show_bug.cgi?id=165505

> However, with 0.2 I was able to fix some problems using just a text
> editor and the RFC.  I have looked in 0.3a1's data file, and I reckon
> any problems would be much harder to fix.  So it seems to me that
> 0.3a1 needs to be much more stable than 0.2 to give the same level of
> confidence.  I don't think we can be certain of this yet.

0.3a1(+) is able to handle *.ics files (I don't store my data in
storage.sdb but in a local "remote" calendar) though there are some
annoying bugs (f.e.
<https://bugzilla.mozilla.org/show_bug.cgi?id=317135>, the patch exists
but you have to apply it yourself).

> Using the alpha, but keeping ICS backups might be a good compromise.

0.3a1(+) backups *.ics calendar files automatically on startup, look
into the backupData folder of Sunbird profile.

--
Ilja
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Christopher Jahn
In reply to this post by Simon Paquet
Simon Paquet <[hidden email]> wrote in news:1133637235.96
@sipaq.newsoffice.de:

> Christopher Jahn wrote on 03.12.2005:
>
>>> How stable is the new Sunbird alpha.
>>> I would like to use it but all my 'stuff' is on 0.2 and I don't really
>>> want to lose it.
>>
>> Stick with 0.2 for now.
>
> Any reasons for that suggestion?
>

I'm a responsible person.

> IMO 0.3a1 is definitely superior to 0.2. It is faster at least as stable
> and has a lot of annoying bugs fixed that plagues 0.2. It has some new
> bugs, sure, but no real showstopper IMO.
>

Then you've been lucky.  There are cases of calenders being completed
deleted.  0.3a1 is an *alpha*.  When it glitches, there are absolutely no
gaurantees that your data will survive.  Now that's fine if A) you only
have one or two small calenders or B) you are only playing around with it
to see what it can do.  

It's an *alpha*.  That alone is enough to throw up warning signs.  The
developers are not confident enough in its stability to label it as a beta.  
It's certainly not mature enough to be a release candidate; and that comes
AFTER it's bumped up to beta.

It's dreadfully irresponsible to recommend it to anyone intending on using
it for any important purpose.  Or at least to do so without saying: "BTW,
you could lose ALL your DATA.  If you absolutely need to rely on it, don't
use this".

This is not the first time you've made this boner mistake, Simon.  It's not
helpful to get people to use a product that is simply not ready for
primetime, only to have them come back in a week screaming about lost
schedules.


--
}:-)       Christopher Jahn
{:-(       http://home.comcast.net/~xjahn/Main.html

I am Zorkon, this is my brother Zenon, and this is my other
brother Zenon.
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Joey Minta
Christopher Jahn wrote:
> Then you've been lucky.  There are cases of calenders being completed
> deleted.  0.3a1 is an *alpha*.  When it glitches, there are absolutely no
> gaurantees that your data will survive.  
I don't think I've heard of a case like this.  If there is one, I
certainly hope a bug has been filed on it, but I don't recall seeing one
of this severity.

> It's an *alpha*.  That alone is enough to throw up warning signs.  The
> developers are not confident enough in its stability to label it as a beta.  
> It's certainly not mature enough to be a release candidate; and that comes
> AFTER it's bumped up to beta.
Sunbird 0.3beta is being reserved for a version that is essentially
feature-complete for 0.3.  The current version an alpha because there's
a lot more we want to do/implement before Sunbird 0.3.

> It's dreadfully irresponsible to recommend it to anyone intending on using
> it for any important purpose.  Or at least to do so without saying: "BTW,
> you could lose ALL your DATA.  If you absolutely need to rely on it, don't
> use this".
No software is perfect.  Sunbird 0.2 can crash and/or mangle your data
(see for instance bug 248884 which, unless you happen to find someone
who knows a fair bit about ics files, effectively renders your data
useless).  Sunbird 0.3a1 has some of these problems, too.  It's not
until we reach Sunbird 1.0 that we'll really feel confident that user's
can store their data safely.

> This is not the first time you've made this boner mistake, Simon.  It's not
> helpful to get people to use a product that is simply not ready for
> primetime, only to have them come back in a week screaming about lost
> schedules.

Let's tone down the language a bit here.  Simon was one of the people
integrally involved in deciding when Sunbird 0.3a1 was ready to be
released.  He gets emails about every bug filed regarding Sunbird (both
0.2 and 0.3a1), and hence would be one of the people I trust most to
give an appraisal of their respective stabilities.

Yes, Sunbird 0.3a1 has problems.  Many of these are detailed in the
release notes.  Sunbird 0.2 has problems as well.  For ultimate
stability, you are probably correct that 0.2 may be a better way to go
(especially if you use alarms a lot).  Simply throwing Sunbird 0.3a1 out
as unusable though, isn't the right way to go either.  Rod's suggestion
of Sunbird 0.3a1 with regular backups does seem like a good compromise.
(Then again, regular backups are always a good idea.)  Importing your
0.2 calendars into 0.3a1, rather than pointing the program directly at
them, can also help reduce the risk.

-Joey
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Christopher Jahn
Joey Minta <[hidden email]> wrote in news:dmt8s0$[hidden email]:

>  Importing your
> 0.2 calendars into 0.3a1, rather than pointing the program directly at
> them, can also help reduce the risk.

FIne for a test case, but not those of us who have an implementation of
0.2.

Jeez, you guys piss me off.  No wonder this project is lagging so far
behind.

--
}:-)       Christopher Jahn
{:-(       http://home.comcast.net/~xjahn/Main.html

Ooooooeee! They sure serve up them four-year-olds good, don't
they?
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Dex-5
In reply to this post by Joey Minta
    I thought Christopher was just being cautious and considerate by
informing Jim to be wary of 0.3a1!

    Cheers,
                Dex


"Joey Minta" <[hidden email]> wrote in message
news:dmt8s0$[hidden email]...

> Christopher Jahn wrote:
>> Then you've been lucky.  There are cases of calenders being completed
>> deleted.  0.3a1 is an *alpha*.  When it glitches, there are absolutely no
>> gaurantees that your data will survive.
> I don't think I've heard of a case like this.  If there is one, I
> certainly hope a bug has been filed on it, but I don't recall seeing one
> of this severity.
>
>> It's an *alpha*.  That alone is enough to throw up warning signs.  The
>> developers are not confident enough in its stability to label it as a
>> beta.  It's certainly not mature enough to be a release candidate; and
>> that comes AFTER it's bumped up to beta.
> Sunbird 0.3beta is being reserved for a version that is essentially
> feature-complete for 0.3.  The current version an alpha because there's a
> lot more we want to do/implement before Sunbird 0.3.
>
>> It's dreadfully irresponsible to recommend it to anyone intending on
>> using it for any important purpose.  Or at least to do so without saying:
>> "BTW, you could lose ALL your DATA.  If you absolutely need to rely on
>> it, don't use this".
> No software is perfect.  Sunbird 0.2 can crash and/or mangle your data
> (see for instance bug 248884 which, unless you happen to find someone who
> knows a fair bit about ics files, effectively renders your data useless).
> Sunbird 0.3a1 has some of these problems, too.  It's not until we reach
> Sunbird 1.0 that we'll really feel confident that user's can store their
> data safely.
>
>> This is not the first time you've made this boner mistake, Simon.  It's
>> not helpful to get people to use a product that is simply not ready for
>> primetime, only to have them come back in a week screaming about lost
>> schedules.
>
> Let's tone down the language a bit here.  Simon was one of the people
> integrally involved in deciding when Sunbird 0.3a1 was ready to be
> released.  He gets emails about every bug filed regarding Sunbird (both
> 0.2 and 0.3a1), and hence would be one of the people I trust most to give
> an appraisal of their respective stabilities.
>
> Yes, Sunbird 0.3a1 has problems.  Many of these are detailed in the
> release notes.  Sunbird 0.2 has problems as well.  For ultimate stability,
> you are probably correct that 0.2 may be a better way to go (especially if
> you use alarms a lot).  Simply throwing Sunbird 0.3a1 out as unusable
> though, isn't the right way to go either.  Rod's suggestion of Sunbird
> 0.3a1 with regular backups does seem like a good compromise. (Then again,
> regular backups are always a good idea.)  Importing your 0.2 calendars
> into 0.3a1, rather than pointing the program directly at them, can also
> help reduce the risk.
>
> -Joey


_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Simon Paquet
In reply to this post by Christopher Jahn
Christopher Jahn wrote on 03.12.2005:

>>>> How stable is the new Sunbird alpha.
>>>> I would like to use it but all my 'stuff' is on 0.2 and I don't
>>>> really want to lose it.
>>>
>>> Stick with 0.2 for now.
>>
>> Any reasons for that suggestion?
>
> I'm a responsible person.

I don't call spreading misinformation or FUD responsible behaviour.
But YMMV there...

>> IMO 0.3a1 is definitely superior to 0.2. It is faster at least as
>> stable and has a lot of annoying bugs fixed that plagues 0.2. It has
>> some new bugs, sure, but no real showstopper IMO.
>>
> Then you've been lucky.  There are cases of calenders being completed
> deleted.

I haven't seen a single bug describing such a case. If there are such
cases, then why are you rambling about it, instead of doing the
responsible thing and file a bug about it, so it can be fixed.

Also 0.3a1 creates backups of remote calendar files automatically on
every startup, so at least this information is unlikely to be deleted
completely.

> 0.3a1 is an *alpha*.

Yes. But that has nothing to do with the initial question.

> When it glitches, there are absolutely no gaurantees that your data
> will survive.

That's the same for every other piece of software on this planet (not
counting software for satellites or nuclear power plants). When they
glitch/crash/hang/whatever you depend on the programmers, whether they
implmented some fallback or not.

> It's an *alpha*. That alone is enough to throw up warning signs. The
> developers are not confident enough in its stability to label it as a
> beta. It's certainly not mature enough to be a release candidate; and
> that comes AFTER it's bumped up to beta.

I'm sorry to be so blunt, but that is complete and utter bullshit!

0.3a1 is called an alpha because it does not fulfill everything that we
want to see in a final 0.3 release. And the beta is supposed to be
feature complete, so we can't call it a beta.

0.3a1 is very stable at this stage, IMO more stable than 0.2 and it will
probably also be more stabel than 0.3a2, when it comes out, because the
move to the new view-code will certainly introduce some regressions.

> It's dreadfully irresponsible to recommend it to anyone intending on
> using it for any important purpose.  Or at least to do so without
> saying: "BTW, you could lose ALL your DATA.  If you absolutely need to
> rely on it, don't use this".

Your argumentation is also flawed because you recommended an 0.2
release.
I mean, come on, you don't want to recommend 0.3a1 because it's an alpha
but you are recommending the 0.2 release without a doubt.

> This is not the first time you've made this boner mistake, Simon.
> It's not helpful to get people to use a product that is simply not
> ready for primetime, only to have them come back in a week screaming
> about lost schedules.

You are right about one thing here. Sunbird is not ready for primetime.
That's why it is at 0.3a1 and not at 1.0. Everybody wanting to use
Sunbird should be aware of that fact. I won't reply to your ad hominem
attacks. It's probably better that everyone here can see you
disqualifying yourself here instead of me doing that for you.

--
Simon Paquet
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Simon Paquet
In reply to this post by Christopher Jahn
Christopher Jahn wrote on 04.12.2005:

>> Importing your 0.2 calendars into 0.3a1, rather than pointing the
>> program directly at them, can also help reduce the risk.
>
> Fine for a test case, but not those of us who have an implementation of
> 0.2.
>
> Jeez, you guys piss me off.  No wonder this project is lagging so far
> behind.

I guess I have enough of your bad behaviour. See you in my killfile!


--
Simon Paquet
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

googlegroups.2005
wow ... a heady discourse. i have learnt 2 things:
1/ You guys are normal developers: you demand high levels of technical
excellence from your peers to the extent of indignance, and you are
ready to man the barricades to defend against technical criticism with
passion and strong language.  Yep, everythings pretty normal here ...
2/ Sounds like 3a1 is in fact ready for a personal run, in preferrence
to 0.2
Yes, 0.x things are almost always non-production, but there are some
libraries which have resisted moving to 1.x for unknown reasons that we
are using in production code, ie: cewolf is now at 0.12.x! (talk about
living on the edge! ooeer!)
Cheers and rage on ... what's a forum without occasional rants and
flames?
Rob

_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar
Reply | Threaded
Open this post in threaded view
|

Re: Sunbird 0.3a1

Simon Paquet
Ribs wrote on 05.12.2005:

> wow ... a heady discourse. i have learnt 2 things:
>
> 1/ You guys are normal developers: you demand high levels of technical
> excellence from your peers to the extent of indignance,

No. We don't demand high levels of technical excellence from our peers.
What I (I can only speak for myself here) demand is, that anyone who is
criticizing the project or the software does so in a constructive way.

It's ok to say "Sunbird 0.3a1 is not a good release, because X, Y and Z
do not work" as long as the same person who is raising these issues is
willing to file bugs about X, Y and Z or at least describes these issues
in a such a way that one of the developers can reproduce the issues and
file the bugs themselves.

> and you are ready to man the barricades to defend against technical
> criticism with passion and strong language. Yep, everythings pretty
> normal here ...

Did you read the discussion here at all? Strong language was only used
by someone from the developer community, when he was personally attacked
with at least the same strong language.

Right now we pride ourselves that discussions and arguments are
presented
in a cool and constructive way. And at least for myself I can say that
I will continue to do so.

> Yes, 0.x things are almost always non-production, but there are some
> libraries which have resisted moving to 1.x for unknown reasons that we
> are using in production code, ie: cewolf is now at 0.12.x! (talk about
> living on the edge! ooeer!)

We'll do our very best to not get into 0.12 or 0.18 releases.

--
Simon Paquet
_______________________________________________
mozilla-calendar mailing list
[hidden email]
http://mail.mozilla.org/listinfo/mozilla-calendar