FWIW, OOo itself also now has...
* spellchecker (hunspell backend) and thesaurus
* email mailmerge feature (pymailmerge)
For Metrically equivalent fonts just...
* grab the Liberation fonts
For "Macro Migration"...
* grab the ooo-build vba implementation, which is a superior approach over the StarOffice mess, ships with most distro OOo variants.
gcj support in HEAD OOo is simply using gij as a bytecode interpreter. Currently I've a trivial patch pending commit to use gcj-dbtool and gcj during buildtime to beforetime compile the java jars used during the build process itself. So during the build process I make use of the gcj native code creation cleverness, but not during runtime as yet.
This is the sort of niftyness which gcj-dbtool and gij enable that I'm talking about.
Yes, been having great success. Since my original post to the gcj mailing lists we've gotten far more working. Crucially the nifty gui access-like front end uses hsqldb and with hsqldb 1.8.0RC9 that now works with gcj. Nearly everything now builds with gcj, e.g. the wizards and accessibility bridge build with gcj 4.0.0. I'm confident that we'll have the wizards actually working properly in week or two.
There is such a tool. The file format is a number of xml streams and binary graphics (if there are any) zipped together. So simply unzip the file and edit the uncompressed resultant xml files by hand, rezip the files together and OOo will be perfectly happy with the document.
There is additionally a single stream "flat mode" xml import/export option which you can use instead of the normal compressed multistream mode, which should be documented somewhere around http://xml.openoffice.org/. In that case a single stream is written and not compressed. Its bigger and slower than the normal mechanism.
Install file, the solver is a collection of prebuilt object files and similiar development stuff to speed up development, not required (or desirable) for casual user install.
FWIW, the next version after 5.2 will have an xml file format which is of course documented (pdf) at xml.openoffice.org.
So anyone can create an import filter to import the new formats. Equally the xml format doubles as an api so creating import filters for OpenOffice is trivial enough
C. (I work for them, so I'm completely biased, but don't speak for them, etc etc.)
I work on the {Star|Open}Office msword import and export stuff. And in the latest versions we do reliably handle fastsaved documents. Get a shiny development interim milestone from openoffice.org and give it a whirl for importing word documents, it laughs in the face of fastsave documents.
Anything that doesn't work is a bug and you can submit it as such through openoffice.org's bugzilla thingy, we can always do with troublesome documents.
I made a patch for xplanet at one stage from Robert Gray's original projection code to draw a Fuller Projection, stuck that standard world map from livingplanet through it to get this pretty neat sample output.
If you can reproduce it with a more recent version then submit a sample document which causes this to happen to OpenOffice's writer component's bug
tracker.
While Word and all the office formats and a large set of windows programs use ole2 storage there are a number of libraries and packages which can read that format. libole2 comes to mind, along with of course the implementation in OpenOffice
No the 2000 file format is backwards compatable with a 97 format reader. I just explicitly said that it can read 2000 or otherwise the assumption would have been that wv cannot read 2000 document. Also FWIW wvWare is GPLed.
While there were bits here and there of the spec that were not exactly as things were in reality I would not characterise the little bits of data dumping here and there to clarify the intent of the format as reverse engineering
The StarOffice source is the same as the OpenOffice source, if you read the openoffice.org documents you can see that the catch is that there are a few bits which didn't belong to sun, those bits cannot be released as they are not owned by Sun, they are being replaced with proper GPLed components.
Theres not a thing anyone can do about that. When you run StarOffice and see all the "components licenced from x,y and z", those are the bits that seperate the old StarOffice 5.2 from the current version.
There isn't a problem to be solved. All else being equal the system is biased towards accumulating votes behind the candidates whose ballot box ranking, i.e. popularity is highest. It would be incorrect to remove Gore before Nader, Gore was more popular than Nader, so Nader must be knocked out. The thing is that once its given than Nader cannot win, you tell those voters who wanted him, that "well you can't have Nader, who would you like instead ?"
Look at your initial numbers, nader is the most unpopular first count candidate. The system is not a mechanism to cause unpopular candidates to get elected, its a system to allow voters who don't get what they want to get their second preference.
There are a number of systems with selections from lists of candidates, various versions of proportional representation, and the truly awesome Single Transferrable Vote in use in Ireland, which really has to be seen in practice for entertaining election counts. Check out the algorithm
Usually yes, but the point is that the site was updated today with the new information as to building it and other documentation, and that is not in googles cache yet.
Hold on two seconds here, less insulting them and needless aggravation. Sane rational dialogue usually works best in these cases, you leaping into the fray wielding some sort of Slashdot Axe is not actually going to help matters, and in fact will hamper future attempts to resolve the situation.
A lot of companies are still new to all of this and need to be convinced, pissing on someones head is not known for its persuasive power. Perhaps you should leave the diplomatic stuff to others. Anyhow how long was it before you released the slash code ?, and where are the other fine coding achievements and submissions to help projects that entitles you to way that we about the place
We all define valuable tools and projects differently, and our needs
might be a bit different then yours
Basically the company was not in a helpful mode and panicked when they saw that their software could be dispense with. They are quite possibly on very weak ground defining their XOR as vital IP, but the way to resolve it is not to send the endless screaming hordes waving placards to their front door, methinks its some quiet emails and representations from sane
parties. These days we could ask redhat or even IBM to phone them and suggest that they help negotiate the problem away.
Bah, rein back on the unilateral action there, whatever about think before you email a quick response, surely theres a lot to be said for think very very carefully before you dangerously rabble rouse editorialise.
This is really a developer issue as to how to arrange their code. I can hardly see the relevence of placing something that will requires a bit of thought to solve on slashdot.
And what was it with the leadin about conspiracies. There is nothing conspiratorial about this, just some historical hiccups.
Im from a fart sized EU member state and I see it as the other way around. We get a commision seat, veto power and a voting system which gives small countries an inordinate amount of power out of all proportion or population size, i.e. if Finland Denmark and Ireland vote together they easily outvote Germany despite having a combined population of about an eighth or Germany.
Most of the small countries have no power anyway, and they are just fooling themselves if they think they have some sort of influence of what happens to them.
Anyhow Switzerland surely enjoys the stability of living in a Europe which is not concerned with intranational rivalries, and the EFTA thing has surely benefited them, so you could consider them as taking the benefits of an EU but not helping one whit, and then smugly wandering about the place preening themselves on their independance and self reliance. Its all in how you look at it.
PathFinder worked unlike last years fiasco and it had airbags. While its voodoo engineering it can't hurt to start strapping airbags onto all nasa's stuff. Airbags onto the side of the shuttle, stick a few around the international space station, just in case we need to bounce it off something.
As this is GPLed code SO is the largest free software contribution, with the word open source used to let you know that its bigger than the pretty much free but opensource mozilla and friends like Plan9 etc, if you mean what the largest previous was then we'd have to run around doing line counts on various programs. The linux kernel is large (though still less than have the size of SO by my incredibly rough count), but is it a "single contribution" as it grew over time piece by piece?, you'd really have to compare against something where some{one|company} freed a body of code in one fell swoop.
I don't thing the GNU project would qualify in this little verbiage contest as its multiply authored as well and not really a single entity as its distinctly lot of programs.
Anyhow I reckon theres a real drop off to number 2.
Only a crass idiot ever choose an os for its underground (hawk!, spit!) and tiny installed base. Heading to BeOS from Linux misses the point entirely.
X used to never crash!, window managers have been falling over for years. Its the source and the freeness to hack the crap out of that said source, stability and so forth is purely icing. XFree86 has had bugs just like everything else, heres the thing though, if you wanted you could fix it, hell if you hadn't the skills you could try to identify when it would happen and help the developers track it down, if you were really into it you could pay a developer to sort it out for you. Even if the original writers had no interest, you could get anyone skilled to do it.
The increased popularity of Linux is great, the naysayers can just bugger off. A strong profile for linux is a strong profile for releasing source, without linux you wouldn't have just gotten 5.3 million lines of source of StarOffice into your laps, what does a little spin to make this announcement accessable to journalists and ordinary investors matter, its the source. And its bloody well GPLed, do you know you awesomely unlikely that is. Its great.
Granted I think that binary only modules for the kernel and so forth are nowhere as good as the real thing, but its a stepping stone along the road for companies to try the waters
This is why Im in this game. Eight years, Ive loved this linux GNU thing. Heres a thought though, how many of you have actually contributed something. Sent in a patch or wrote some code, thats the community, not the gripers about how popular it has become, what kind of mad talk is that.
No matter how it all works out, with linux ventures dissappearing into the sea, or companies bailing out of free software the deeds are done, and the code is there for all or us to play with to our hearts content, and thats what its all about. Hmm, port SO to OpenBSD, go ahead, excellent hack, but its the popularity of linux which you so fear which has created that opportunity. Popularity is a tool, take advantage of it
Slashdot Mirror
FWIW, OOo itself also now has...
* spellchecker (hunspell backend) and thesaurus
* email mailmerge feature (pymailmerge)
For Metrically equivalent fonts just...
* grab the Liberation fonts
For "Macro Migration"...
* grab the ooo-build vba implementation, which is a superior approach over the StarOffice mess, ships with most distro OOo variants.
gcj support in HEAD OOo is simply using gij as a bytecode interpreter. Currently I've a trivial patch pending commit to use gcj-dbtool and gcj during buildtime to beforetime compile the java jars used during the build process itself. So during the build process I make use of the gcj native code creation cleverness, but not during runtime as yet. This is the sort of niftyness which gcj-dbtool and gij enable that I'm talking about.
Yes, been having great success. Since my original post to the gcj mailing lists we've gotten far more working. Crucially the nifty gui access-like front end uses hsqldb and with hsqldb 1.8.0RC9 that now works with gcj. Nearly everything now builds with gcj, e.g. the wizards and accessibility bridge build with gcj 4.0.0. I'm confident that we'll have the wizards actually working properly in week or two.
There is such a tool. The file format is a number of xml streams and binary graphics (if there are any) zipped together. So simply unzip the file and edit the uncompressed resultant xml files by hand, rezip the files together and OOo will be perfectly happy with the document.
There is additionally a single stream "flat mode" xml import/export option which you can use instead of the normal compressed multistream mode, which should be documented somewhere around http://xml.openoffice.org/. In that case a single stream is written and not compressed. Its bigger and slower than the normal mechanism.
C.
Install file, the solver is a collection of prebuilt object files and similiar development stuff to speed up development, not required (or desirable) for casual user install.
Star/OpenOffice opens office 95 documents.
And for the record is is *not* written in java ok :-)
What might be useful is to read how StarOffice, did their unicode and internationalization changes to an existing large code base at sun.com
C.
FWIW, the next version after 5.2 will have an xml file format which is of course documented (pdf) at xml.openoffice.org.
So anyone can create an import filter to import the new formats. Equally the xml format doubles as an api so creating import filters for OpenOffice is trivial enough
C. (I work for them, so I'm completely biased, but don't speak for them, etc etc.)
I work on the {Star|Open}Office msword import and export stuff. And in the latest versions we do reliably handle fastsaved documents. Get a shiny development interim milestone from openoffice.org and give it a whirl for importing word documents, it laughs in the face of fastsave documents. Anything that doesn't work is a bug and you can submit it as such through openoffice.org's bugzilla thingy, we can always do with troublesome documents.
C.
C.
C.
While there were bits here and there of the spec that were not exactly as things were in reality I would not characterise the little bits of data dumping here and there to clarify the intent of the format as reverse engineering
C.
Theres not a thing anyone can do about that. When you run StarOffice and see all the "components licenced from x,y and z", those are the bits that seperate the old StarOffice 5.2 from the current version.
C.
Look at your initial numbers, nader is the most unpopular first count candidate. The system is not a mechanism to cause unpopular candidates to get elected, its a system to allow voters who don't get what they want to get their second preference.
There are a number of systems with selections from lists of candidates, various versions of proportional representation, and the truly awesome Single Transferrable Vote in use in Ireland, which really has to be seen in practice for entertaining election counts. Check out the algorithm
C.
C.
C.
A lot of companies are still new to all of this and need to be convinced, pissing on someones head is not known for its persuasive power. Perhaps you should leave the diplomatic stuff to others. Anyhow how long was it before you released the slash code ?, and where are the other fine coding achievements and submissions to help projects that entitles you to way that we about the place We all define valuable tools and projects differently, and our needs might be a bit different then yours
Basically the company was not in a helpful mode and panicked when they saw that their software could be dispense with. They are quite possibly on very weak ground defining their XOR as vital IP, but the way to resolve it is not to send the endless screaming hordes waving placards to their front door, methinks its some quiet emails and representations from sane parties. These days we could ask redhat or even IBM to phone them and suggest that they help negotiate the problem away.
Bah, rein back on the unilateral action there, whatever about think before you email a quick response, surely theres a lot to be said for think very very carefully before you dangerously rabble rouse editorialise.
C.
And what was it with the leadin about conspiracies. There is nothing conspiratorial about this, just some historical hiccups.
C.
Most of the small countries have no power anyway, and they are just fooling themselves if they think they have some sort of influence of what happens to them.
Anyhow Switzerland surely enjoys the stability of living in a Europe which is not concerned with intranational rivalries, and the EFTA thing has surely benefited them, so you could consider them as taking the benefits of an EU but not helping one whit, and then smugly wandering about the place preening themselves on their independance and self reliance. Its all in how you look at it.
Anyhow Austria isn't a large state.
C.
I don't thing the GNU project would qualify in this little verbiage contest as its multiply authored as well and not really a single entity as its distinctly lot of programs.
Anyhow I reckon theres a real drop off to number 2.
C.
X used to never crash!, window managers have been falling over for years. Its the source and the freeness to hack the crap out of that said source, stability and so forth is purely icing. XFree86 has had bugs just like everything else, heres the thing though, if you wanted you could fix it, hell if you hadn't the skills you could try to identify when it would happen and help the developers track it down, if you were really into it you could pay a developer to sort it out for you. Even if the original writers had no interest, you could get anyone skilled to do it.
The increased popularity of Linux is great, the naysayers can just bugger off. A strong profile for linux is a strong profile for releasing source, without linux you wouldn't have just gotten 5.3 million lines of source of StarOffice into your laps, what does a little spin to make this announcement accessable to journalists and ordinary investors matter, its the source. And its bloody well GPLed, do you know you awesomely unlikely that is. Its great.
Granted I think that binary only modules for the kernel and so forth are nowhere as good as the real thing, but its a stepping stone along the road for companies to try the waters
This is why Im in this game. Eight years, Ive loved this linux GNU thing. Heres a thought though, how many of you have actually contributed something. Sent in a patch or wrote some code, thats the community, not the gripers about how popular it has become, what kind of mad talk is that.
No matter how it all works out, with linux ventures dissappearing into the sea, or companies bailing out of free software the deeds are done, and the code is there for all or us to play with to our hearts content, and thats what its all about. Hmm, port SO to OpenBSD, go ahead, excellent hack, but its the popularity of linux which you so fear which has created that opportunity. Popularity is a tool, take advantage of it
C.
disclosure: I work for SO