[Lazarus] Loading old project into Lazarus_fixes causes lock-up...

Martin lazarus at mfriebe.de
Mon Nov 29 12:02:14 CET 2010


On 29/11/2010 07:17, Graeme Geldenhuys wrote:
> Yes, and consider the other side of the coin. It will also add many many
> new bugs. I also use 0.9.28.3 in a VM session for our Windows development,
> and testing of our projects. I have no problems at the moment, but I don't
> dare update that VM either, because I'm pretty sure if I do, it will have
> lots of new problems I don't need now.
>
> The Golden Rule of using Lazarus
> --------------------------------
> If you find a version or revision that works for you - stick with it for as
> long as possible! Do not be tempted to upgrade, because what works today,
> will most likely be broken tomorrow. Mantis reports of proof of that.
> [sorry if this offends some, but the truth hurts sometimes]
>
<pretext: before starting a list => read the end of my mail>

The truth in itself is not necessarily hurtful (or as hurtful).  It 
often is the way how it is told or what is presented as the truth.
All of what you say is true, it is part of the truth. (Maybe It is your 
pick of the truth). It is  not "the truth" as in the complete  truth. 
There is much much more to the truth.
Every truth has good and bad in it. If anyone chooses to tell the bad 
points only, and then announce it as the truth, well yes, that does hurt.

And btw quote: "version or revision". As for revisions it is clearly 
said that they are not tested at all, so on this part you are absolutely 
right. But it is not considered a problem.
There is gratitude to everyone who uses snapshots or SVN and helps the 
development by pointing out the issues found.
But using snapshots is meant to be experimental, it is meant to be one 
way of helping the development.
It is *not* meant to be used for development. If you do use the 
snapshots outside there purpose, and you complain that they do not work 
for what you attempt, then sorry, but they are not indented to work for 
that purpose.

And to be clear. That does not mean that "versions" (released versions) 
are 100% absolutely fault free. How could a small team of people 
(working in there spare time only) archive, what no one (no team however 
big or small) has archived before?

And what about mantis? Go to other bugtrackers of other software (of 
reasonable size). You will find with many, they are neither empty, nor 
free of regression-reports.

Another bit of the truth is that Lazarus is not different from other 
software. The truth (or rather another part of it) is that no Software 
is infallible.


Apart from this: Yes there are certain areas where Lazarus may have 
shortcomings  compared to the average, but so there are areas where 
Lazarus is above the average. If anyone want to start a list, I will 
*not* enter this senseless battle, as both sides can create huge lists, 
simply pointing out what is already known, and what has been said many 
many times before.

If anyone wants to contribute, and help improving them => please do

Martin





More information about the Lazarus mailing list