incubator-ooo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Weir <>
Subject Re: Some stats and observations on OpenOffice upgrades
Date Wed, 26 Sep 2012 18:55:44 GMT
On Wed, Sep 26, 2012 at 9:35 AM, Alexander Thurgood
<> wrote:
> Le 26/09/12 07:11, Reizinger Zoltán a écrit :
> Hi all,
> Many public administrations in France, e.g. schools/colleges, town/city
> halls that would have had OOo 3.3.x have not upgraded en masse : either
> they have switched / are slowly switching to LO, or else they are
> waiting for a feature complete version of AOOo on a par with what they
> had in 3.3.x (or even 3.2.x in some cases) to come along.
> Currently, as Zoltan has pointed out, that is not the case for the Base
> module, which shows problems in ODBC/JDBC crashes/freezes, Reports and
> Mailmerge features. The French administration makes fairly heavy use of
> the database and mailmerge features, if they aren't working properly,
> whether it be in AOO or LO, then obviously the admins are loathe to
> switch, despite any security concerns that may be apparent. If you can
> lockdown your security without losing functionality on an old version,
> why change to something that might be more secure, but only gives you
> three quarters of what you need ?
>> Some features not supported in AOO 3.4.x especially, the report builder
>> not developed any more,
>> and Base users meet problems with reports containing charts.
>> For this reason they after upgrade to AOO 3.4.x downgrade back az least
>> one computer in companies
>> to OOo 3.3 for report creating, and not care about update warning.
>> And waiting for solution, and we can not provide them real solution,
>> It is from forum posts (LibO no solutions for them, because their report
>> builder has different bugs,
>> which prevent from daily work.)
> Yes, unfortunately, _both_ LO and AOO in their current incarnations
> suffer from some serious usability issues with regard to Base, although
> the rate of bug fixing with regard to these issues is higher (at least
> in appearance from my following of the various bug reports) on LO than
> AOO. The biggest problem that I see with LO as regards Base is that the
> frenetic feature development in the other modules often comes at the
> expense of increased instability in Base, due to incompatible code
> changes, simple overwrites in git, or forgetfulness to adapt the changes
> to those parts of the code that will also affect Base.
> How many people are regularly working on fixing/maintaining the code in
> Base in AOO ?
> In the LO project there is mainly only 1 person at the moment in his
> free time (although at least one other person is also starting to get
> involved). To me, this shows clearly that past history, division of
> labour, and incompatible code licensing schemes have not helped either
> project. It also shows that Base development is not considered rewarding
> enough by most would-be contributors. I don't remember how many people
> were working on Base within Sun, but I seem to recall it was mainly 2 or
> 3 developers, plus a QA person (I could be wrong, it is just that I
> never met any of the others online or in the bug reports)
> The impact of the current state of affairs plays on people's fears - why
> move to either LO or AOO when you can stick with OOo 3.3.x and continue
> as before ? Both projects suffer from this image issue and the users
> ultimately, are the one's who suffer. Yes, it might be free, but if it
> doesn't work in the way it used to, why use it at all ?
> These fears are those that need to be allayed, and they are not
> unfounded. If the functionality of a user's choice of software goes down
> from one release to the next, then that user will drop that shiny new
> version in favour of its tried and trusted workhorse.

This may all be true, but it doesn't explain the data we're seeing.
If AOO was unacceptable to the remaining OOo 3.3 users then we would
expect to see a large upgrade count when AOO was first released,
followed by the upgrade rate trailing to nothing.  Eventually you
reach the point where everyone who finds AOO acceptable has already
upgraded.   But we don't see that.  We see, after the initial surge, a
steady upgrade rate of around 20,000 OOo 3.3.0 upgrades/day, even 4
months after AOO 3.4 was released.   The odd thing is that this number
is not declining over time.  Maybe we just need to wait longer.

Also note that we're still seeing OOo 2.0 upgrades, even though those
users have been offered OOo 3.2.1 and 3.3 upgrades, as well as AOO 3.4
upgrades, for a couple of years now.


> I'm not saying that the above is the only reason, just one of many
> factors that needs to be considered. If the project wants AOO to be
> adopted, then it has to look to provide the things that were there in
> previous versions of OOo.
> Alex

View raw message