openoffice-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Inge Wallin <i...@lysator.liu.se>
Subject Re: OpenOffice thin client edition - why not?
Date Sun, 27 Jan 2013 13:34:31 GMT
On Sunday, January 27, 2013 03:33:20 Rob Weir wrote:
> On Thu, Jan 24, 2013 at 9:29 AM, Inge Wallin <inge@lysator.liu.se> wrote:
> > On Wednesday, January 23, 2013 21:33:04 Rob Weir wrote:
> >> On Wed, Jan 23, 2013 at 8:06 AM, Inge Wallin <inge@lysator.liu.se> wrote:
> >> > On Friday, January 18, 2013 15:21:01 Ian Lynch wrote:
> >> >> On 18 January 2013 13:18, Fernando Cassia <fcassia@gmail.com>
wrote:
> >> >> > On Fri, Jan 18, 2013 at 9:43 AM, Ian Lynch <ianrlynch@gmail.com>

wrote:
> >> >> >> What we really need is a cloud version of AOO like Google
Docs.
> >> >> > 
> >> >> > We don´t *need* ONE thing. That´s the beauty of open source,
´we´
> >> >> > could do *several* things.
> >> >> 
> >> >> Well yes, but it is more efficient to do one thing that covers many
> >> >> needs rather than try and do many things with not enough resource.
> >> >> 
> >> >> > I for one don´t ´need´ an AJAX / HTML5 version of AOO... GDocs
is
> >> >> > fine...
> >> >> 
> >> >> A lot of people would say yes but GDocs is not open source.
> >> >> Some people would say MS Office is fine and others would say Koffice.
> >> >> Question is whether or not we want a long term sustainable project
> >> >> for the community or one that will get more and more marginalised.
> >> > 
> >> > As a side note: While I am happy that KOffice is mentioned now and
> >> > then on this list, I think it would be proper to mention the Calligra
> >> > Suite instead. KOffice is not being developed any more while Calligra
> >> > is running full speed ahead.
> >> 
> >> Hi Inge,
> >> 
> >> Thanks for the reminder.  Getting people to recognize a name change
> >> takes time, and repetition.  We still see on a daily basis people
> >> expressing surprise to learn that OpenOffice is now at Apache.
> >> 
> >> Are you planning to be at the KDE conference in July in Bilbao?   It
> >> might good to have someone from AOO attend.  Aside from the obvious
> >> common interest in ODF, it would be interesting to see if there are
> >> any other opportunities for collaboration.
> > 
> > Yes, I will definitely be in Bilbao unless something very serious
> > happens. I would love to meet you there and talk about collaboration
> > between our projects.
> > 
> > I can see two areas where we could start immediately without much talk:
> > 
> > 1. test documents
> > 
> > Calligra has ~3500 test documents in ODF (odt, ods, odp), MS binary (doc,
> > xls, ppt) and MS xml (docx, xlsx, pptx) formats. I suppose that there is
> > a test suite available for AOO as well.
> 
> Are these documents "from the wild" or documents created specifically
> for tests?  I remember hearing Jos describe a technique for creating
> test documents that sounded interesting.

They are not hand hacked xml but where created specifically for testing the 
respective function by testers employed by Nokia. They where done originally 
for the document viewer application in the Nokia N900 and then later extended 
for the viewer in the Nokia N9.

> I recall Microsoft having a collection of test documents as well, that
> they shared at a plugfest a few years ago.

Must have been one that I didn't attend.  :/  It would be interesting to see 
those documents.

> I have a few "interesting" test documents, but the Symphony test
> documents are IBM-internal right now, since many of them are customer
> files that we may not share.  But if there is interest (across
> projects) in creating a collection of test office documents in several
> formats, then I would investigate to see if there are some that we
> could contribute.

The most "interessting" one we have is created by MS Office.  It's valid ODT 
but not structured the same way as LO/AOO normally does it.  We had to work a 
lot to make that render correctly.  :)

> > The documents that Calligra has access to are structured not only
> > according to format but also to feature, such as pictures, text
> > formatting, graphics (smart art, etc), and so on.
> 
> Excellent.
> 
> > It would be great if we could work to create an even bigger and better
> > database of test documents which covers even more features.
> 
> Yes.  But where to do this?  OASIS is not really set up to do this.
> (and it sounds like it would be best to do more than just ODF test
> documents),  It could be done in the AOO project, but that may make it
> difficult (politically) for some to contribute.  Without arguing the
> reasons for that view, I think it is (sadly) current reality.  Other
> choices might be the ODF Toolkit project (where we have the ODF
> Validator) or OfficeShots (which allows automated online testing).

Or just continue with the KDE repository? Check out:  
svn://anonsvn.kde.org/home/kde/trunk/tests/calligratests
Getting a KDE commit account is much easier than an Apache one.

> > 2. Interoperability and ODF confomance.
> > 
> > It would be good if you could give a high priority to bugs which make
> > interoperability with other ODF suites such as Calligra. In Calligra we
> > already have a special bug category for ODF related bugs and these are
> > always treated with speed and priority.
> 
> We have an XML "product" in Bugzilla where ODF bugs are categorized,
> as well as other XML-related import/export issues in XHTML, XSLT, etc.
>  But if you have a specific set of ODF issues that you think we should
> raise in priority, posting that list to this mailing list would help.
> 
> Regards,
> 
> -Rob
> 
> > What do you say?
> > 
> >         -Inge
> >> 
> >> Regards,
> >> 
> >> -Rob
> >> 
> >> >         -Inge
> >> >> > 
> >> >> > I personally think browser based apps are a pig, and doing apps
in
> >> >> > JScript is insane. I had Chrome open the other day just with GMail
> >> >> > and it was using over 150 MB of RAM...
> >> >> 
> >> >> Not really a big problem with modern multi-gig computers (including
> >> >> future mobile technologies). Less of a problem than stuff that only
> >> >> works on one device or needs a lot of effort to port across
> >> >> multi-devices, operating systems etc. To me open standards are worth
> >> >> paying a bit of a price for in terms of machine resources since the
> >> >> latter continue to grow and get less expensive.
> >> >> 
> >> >> > A thin client virtualized version on the other hand would use
the
> >> >> > PC´s CPU and horsepower and deliver great speed to even to lowest
> >> >> > powered devices.
> >> >> 
> >> >> Assuming you have someone to host it for you. O a global scale that
> >> >> is not trivial to do which is probably why Google with all its
> >> >> resources does what it does.
> >> >> 
> >> >> > But of course, that´s going in a different direction from the
> >> >> > current fad....
> >> >> 
> >> >> Swimming against global trends is not a sensible idea when you have
> >> >> very limited resources and very little time.
> >> >> 
> >> >> > FC

Mime
  • Unnamed multipart/alternative (inline, 7-Bit, 0 bytes)
View raw message