Slashdot Mirror
Opening the Potential of OpenOffice.org
[vmlinuz] writes "O'Reillynet is running an article about 'Opening the potential of OpenOffice.org' which explores how anyone can contribute to argubly one of the most important Open Source projects. The article also discusses the importance of a shorter release cycle."
If something's in beta, people won't want to use it because it just doesn't sound reliable. If it sounds like a stable, final release, people will be more willing to use it, thereby finding the bugs, thereby resulting in bugfixers, which leads to more reliable software.
Personally I prefer LaTeX and send pdf files. That works ok till I am working alone. But if we have to work and interact, keeping track of changes is not the easiest thing to do in LaTeX.
I didn't realize that the OpenOffice project had only 100 developers. Many more will be needed to establish the kind of release schedule mentioned in the article. Interesting stuff. Is this a potential weekness of open source - an inability to attract more developers who will donate their time?
"Me fail English? That's unpossible." - Ralph
You will find variants of Office on almost every Windows PC used in the business and academic worlds, so it's pretty important.
For me, one of the best features of OpenOffice is its ability to export documents as PDFs. Lovely, lovely feature for creating interoperable documents. Yeah, yeah, I know not everyone likes PDFs, but there are PDF readers for most platforms...
EricJoin my mailing list and win a free book
"explores how anyone can contribute to argubly one of the most important Open Source projects."
Not the most important project, but one of them..
I think openoffice is just as important as linux anyway. (This is helped because i dont think of linux as *that* important anyway, being a big bsd fan - but thats a discussion for another time). I think if you want people to switch to an open source operating system you need to take it in steps, making programs like firefox and openoffice (which will run on windows inplace of IE and MS Office) a vital part of the plan. Once you have changed all their apps over to open source versions, you can switch the os and all they will notice is a new look.
Paul
This is a timely and helpful article. As a non-coder I have often wondered how I could help, but except for turning people on to Linux locally, and helping them over the learning hump, I admit I have been lazy and haven't done much.
This article paints a good roadmap for more. Thanks.
C'mon now. One of the most important open source projects in the world? I suppose that assumes that MS Office is one of the most important programs (suite...whatever) in the world? For real?
YES!
The office suite is the one application that keeps people on Windows!
The others were authored by *cough* Google *cough*...
(Weird that, huh?)
If my grammar and spelling are off, I am [distracted/tired/careless] (take your pick)
and? Are we supposed to see your brother as the most important person in the world or something? My brother won't switch to Linux because he doesn't know how to work a command line and we still don't have a decent control panel applet on any Linux distribution worth mentioning. So why isn't that the most important application? When you don't need to operate a shell to configure your computer, then we'll see more defections, not before.
How we know is more important than what we know.
I agree that the OOo guys need to draw a line in the sand soon with 2.0, go gold, if for no othere reason than the current 1.1 is so insainly lacking compared to MS office or OOo2 beta. But just because the number is higher doesnt make it better, want proof, look at Adobe Reader, what can 7 do that 5 can not?
Some basic things like being able to have dotted lines would be nice.
I think it can be argued that, for wide-scale adoption of Linux, the first step will be the wide-scale adoption of OpenOffice over the MS office suite.
After that, switching out the underlying OS becomes transparent (Ok, more transparent for more people).
I guess I subscribe to the idea that a key foothold MS has (at least in the corporate world) is that all of our data is stored in their propietary file formats. Or, in other words, the problem in switching people over isn't that they have to run a MS os, they have to run the MS apps, in particular their office suite. Excel and Word are defacto standards to run a business -- and by extension the MS OS.
It's in that sense that I do think OpenOffice is incredibly important to the OSS world at large. The threat of being a credible (or higher quality, more useful) replacement is higher than with what's happened (and happening) with Firefox vs IE, since IE is also free. MS Office is far from free -- and I think it'll be easier to justify abandoning it because of the cash saved.
If I were MS, I do think OpenOffice is the one OSS project I'd be most nervous about, as it's one of the major threats to the monopoly, and an attack on one of the biggest reasons companies are forced to pay for the MS OS.
BTW, the web browser is probably the other "very important app" for the same reasons, and it's cool that Mozilla Firefox has grown so much. At work it doesn't matter that I choose to run Linux, since I'm running the same web browser as many people who are running Windows (my company is already formally supporting, and recommending, Firefox for internal use). Again though, imagine that IE was an extra app that companies had to pay money for -- I wonder what the Firefox adoption rate would be.
One last thing, it's no surprise that MS has from the beginning to "subvert" the web and web standards. It's all about the formats. I guess they simply arrived way too late to the Web to completely take it over. But I'm sure they know that if they had managed to switch everyone over to ms-propietary-html to surf the web, we'd be paying through the nose for IE and their OS and Office monopoly would be further protected.
I agree that Open Office is one of the most important open source projects. This is because it won't be a Linux derivitive that makes its way onto the desktops of the masses first. It will be open, free applications that can reliably provide the benifit of expensive commercial applications on the *Windows* desktop. A company I work for is interested in an open source "Save to PDF" tool because, well, have you priced Adobe's Acrobat lately? Not cheap. So, they are willing to consider this open source replacement to distribute to the general population. It provides most of the functionality that most of their user base needs and saves them money. The users don't even need to learn anything new. But ask them to swap out their enterprise desktop? Forget about it. If Open Office can get there (and it will *long* before Linux deriviti do), the Corporate World(TM) will open its loving arms.
Bang Logic - Serious Small Business Services
From the point of view of business, there are two fundamental applications:
1) Email
2) Wordprocessing/Self publication.
These are the drivers.
Personally, I see three killer apps from Microsoft (or currently owned by Microsoft) that yet to have equivalents in the open-source world:
- Excel:
The power of the Excel in power-user mode is phenominal. The scalability, programability and calculation abilities of this program are amazing. Open Office does not, as yet, scrape the surface. That OO calc is enough for 90% of all users means that it won't get into businesses where the other 10% need to share data.
- Project
- Visio
I'll bundle these two, as neither are particularly complicated, but the file formats have become defacto standard. Once open source tools can import and export these formats, we'll be able to start displacing them on the desktop.
Norman Cook's Ode to Sl
Maybe someone can create a new GUI for OOo that doesn't like that of Office 97. Sad but true, much open source software clings to GUIs of old closed source designs (Nautilus : MacOS 9, Epiphany : Netscape 4, AbiWord : Word 97 etc.). I wonder when open source developers will make GUIs that are innovatively good, rather than creatively bad (Blender, Grip anyone?).
This is the stupidest argument I've ever heard. Allow me to summarise: Linux isn't broken because Windows is broken. How freakin' braindead is that? Face reality, people use Windows. If you want people to use Linux you have to be better than Windows. You can't say "Windows is broken too" like a child, you have to fix the god damn problem.
How we know is more important than what we know.
Jon Udell has an interesting idea of reinventing the office suite for a networked world. He says it should include "service orientation, peer-to-peer capability, workflow, federated identity, and new ways to query and visualize data." With the source code, someone could develop a system that could improve inter-company communication and collaboration using Open Office. We need to think 21st century.
I understand Abiword and Gnumeric can't replace the entire MS suite, but surely word processing and spreadsheet are the most common office suite applications (except maybe email, which OO doesn't have either.) I certainly don't understand why an integrated bloated "Office Suite" like OO is needed to replace MS Office, when Abiword and Gnumeric seem to me to be doing a much better job right now than OO.
We don't necessarily need a single office suite like OO to replace MS Office. Right now I would support Gnumeric and Abiword.
Penny - plain text accounting
A little tacky, but not as much as your _two_ trashy adverts in you sig. Suck it up.
Writer vs. Word isn't much of a comparison unless you want the collaboration.
Excel vs. Calc I tend to think isn't as big as a lot of people make it out to be. I think there is plenty of room to improve on the OOo side and take the lead. Allowing the use of sometihng like python for macros to produce new functions which can be put in to cells and allow the python to travel with the document would make it easy to produce plug-ins to fix and calculation short comings. As far as supporting MS's file formats and such, I think it's as good as it should really get unless there are a bunch of volunteers that want to work on it. It's a losing battle trying to stay compatible with MS. I'm not a super hard core spreadsheet hacker and I see a lot of people using them just as tables or for column based editing rather than performing a lot of cool calculations.. but that's just my opinion. I mention python becuase you can securely load it if you choose to.
The other apps are pretty close also. The big missing pieces are visio and project.
Now what I think OOo should do is start forking off in a different direction. I'd like to see openpgp integrated so I can sign documents or encrypt them to my colleagues, seemlessly. Upon that kind of framework then start to build collaborative editing. It needs to be made for the 21st century, using web technologies and such. Integrate seemlessly with subversion... MS Office is stagnating, they do big upgrades but lately it has been more UI related than serious function. Look at some of the stuff Lotus has done with Notes, I know it's laughed at by many but that's because they don't understand it, you can do some really really cool stuff with it too.
Performance needs to seriously looked at. I know they hear it a ton and I know they've worked on it a lot. I think it's quite usable once it starts up, there has to be some other things that can be addressed though.
Something else that I know has been hacked around on and talked about and I've seen demos of is further integration to the desktop. In particular, kparts, xpcom and bonobo type objects with some intelligence. As blogs and shit like that continue to grow and there are more and more web based interfaces (it's probably gone beyond "fad" guys..
It's about as lame as putting shameless ads in your sig which isn't a sig.
Thank you for pointing out the obvious, I've been waiting for someone in the Slashdot community to notice this fact: The next MS Office is at LEAST half a decade more advanced than OpenOffice.
I'm not a Microsoft fanboy by any stretch of the word, and I REALLY cringe when some PR bozo starts spouting words like "innovation" to describe their company's products. Having said that, MS Office 12 is the DEFINITION of innovation. Yes, I just cringed at myself, but it's true. If you watch that video, you simply cannot deny the truth in this.
Linux may be one of the more advanced operating systems in the world, and a bunch of open source software may be the best in their field by leaps and bounds. However, OpenOffice is really going to be in trouble, saved only by the fact that it's "free".
In fact, looking at the differences between Office 12 and OO brings up a very obvious point about open source software: VERY few OSS developers are willing to innovate (with notable exceptions, of course). Look at KDE copying Windows, for example, with a start button and all that crap. You find this in most popular OSS projects, they copy the look, feel, and functionality of commercial products without really doing any innovation (yes, that silly term again). This is the biggest weakness of OSS, by far. They look like clones of existing commercial software (with usually just a handful of improvements). This is a weakness because when a commercial company like Microsoft releases a completely redesigned product like Office 12, the OSS competition is immediately cut off at the knees.
Having said all this, I wish the OO.org development community a lot of luck. They've been busily trying to clone what is essentially the 1997 version of MS Office, and perhaps this will be their wake-up call. I'm sure that one of the next releases, perhaps two or three years from now, will be similar to Office 12. But perhaps, given this sudden jolt of REAL competition from Microsoft, they'll design something even BETTER. And if that happens, we will all benefit.
UNIX: A computer user is defined as a programmer. WINDOWS: A computer user is defined as a consumer.
I totally agree. To me, Access is like the swiss army knife of data processing.
.mdb file and you are done.
Have a list to mangle? Shove it in a table or two and run some queries on them.
Want to query to totally unrelated databases that use totally different database servers? Link the tables via ODBC and run queries.
Create a really basic data entry tool? Build a form that feeds a table in literally minutes.
Want to easily move that little form to another computer for someone else to fill in? Just copy that ONE
Plus, if you want to move into something with a real database, well, Access makes a great front end to your full featured database.
I'm not a huge MS fan, but they do have some excellent products. Access, and Exchange/Outlook and SQL Server come to mind.
How does this get modded as interesting? I continue to argue, while being ignored, that this was a mistake for everyone except those that ONLY use a WebBrowser or Email client and neither at the same time. For everyone else, it was better to slim down the Mozilla Suite.
:-).
Anyone ever look at the memory footprint of having FF and TB open at the same time and compare that to the footprint used by the Mozilla Suite? I didn't think so. The former combination is huge when compared to the latter. Why have 2 instances of the GRE open when 1 will do? You know, use it as it was intended to be used!
Just look at my comments below. The 5.1 StarOffice Version ran MUCH MUCH MUCH faster than any one of the OO programs. That was a desktop suite that included everything INCLUDING the kitchen sink.
What we have is crappy coding and a crappy implementation. I hope to God this isn't the way coding is going.
I do agree with you last sentence. "Let's free ourselves of the unwieldy bloat it has given with us" and I'll add " and give us back the StarDivision programmers and the FAST Desktop Environment"
Don't make the all too common and downright stupid assumption that individual packages is always better. It's not. It depends on the user and this user wants the speed back. I have multiple windows open and multiple programs open. I use a spreadsheet with a wordprocessor (at the same time) and I have my email client and my web browers open at the same time.
Give me efficient programming led by someone with vision and hopefully performance in mind.
mod me down for having valid points so that we can continue having assinine comments elevated as insightful or interesting.
Sorry, It WON'T be killer app. For that cause that MOST features of collabration is not needed by common crowd. Yeah, managers and busness people would love that (and I DON'T downplay it's importance), and they are significant, but just a part of market.
This app won't go on Win98, Win2000, etc. So forget it. These people WON'T change their computers for that cause.
Therefore to say that, OO.o could have more commercial sollutions for collabration, AFAIK, there are lot of free ones.
user@ubuntubox:~$ stfu This server is going down for shutdown NOW!
1) Open Office needs to reach the good enough mark. I think it is getting very close to that for a lot of people.
The whole tech support thing is a bit of a non issue - I only know of one person who has called Microsoft for tech support - they weren't very impressed. What mums and dads like better than somebody to ring its somebody they can take the computer to and have them fix it. As more IT professionals become familiar with OpenOffice this sort of support will become available.
Although I can't see OpenOffice in costco in the immediate future - I am willing to bet that you will be able to pick up a copy in most newsagents.
partly true - Openoffice needs to be compatible with the features that are frequently used and is likely to be experienced by that user. The Office compatibility thing only really works if you have everybody using the same version of office in the same environment. It might work that way in other parts of the world but it does not work that way in my area. People very quickly learn what they can and can't send to other people. As long as OpenOffice is as good or better than using a different version of office I think that will be good enough.
Open Office was the prefered name. But it turns out that name is intellectual property of somebody else.
2) Have you actually seen a mum and dad user with a windows installer for the first time. They freeze they get through two maybe three of the questions and give up. It is very intimidating. Essentually what should happen is the user should click once agree to the license (if absolutely necessary) and install (two clicks tops) (maybe with an option for advanced users to tweek the settings before install). The best installer I have used on linux is Synaptic. It is excellent for expert users. If somebody could create a polished version of this that would show only the applications that a normal user would be interested in (by default) and let them install them with a single click I think we might be onto a winner, I agree with the comments about the start menu however
3) There are installers that surpass the ease of use of the windows XP installer. Mepis Linux is my favourite. You put the CD in boot from it - You end up with a fully working desktop... There is an Icon on the desktop that says "Install me" You answer a couple of questions (things like language settings and location) Then the whole lot gets copied to your hardrive. The extra complexity comes from the fact that Mepis allows you to install itself along side other operating systems - which is something people generally want to do... To aid in this task it runs a program very simular to partition Magic - A very nice touch
4) a - these will come as the market matures and demand increases - There are things that nead to be done in order to make these systems popular enough for these chain outlets to stock it.
b -
I really like the idea behind LaTeX -- I've used it for about 5 years now (mostly academically), but I've always had problems when it's come to writing my own classes and packages.
I've found LaTeX is fine to a certain point, but I've also found that as soon as it comes to writing any sort of presentation code, the documentation becomes a lot harder to find and usefully comprehend. I've also had a lot of problems figuring out TeX and LaTeX as languages at all.
There seems to be so much potential for conflict between packages, and the whole thing feels quite flakey. Even in standardly-provided classes, I've lost count of the number of times where I've read things like "it's okay to do this, but make sure you don't do it at the same time as that, or you'll have problems". They're numerous things like not being able to use a table recursively, or having a page environment act like a page 90% of the time, but inexplicably break with the other 10%. Maybe it's something like being able to get the width of object A and being able to set the width of cell B, but not being able to insert the result of one into the parameter for the other.
The need to compile something two, three or four times just to make sure references are all up-to-date seems a little unnecesary and overly-complex in today's world. It might help for compiling large documents without much memory, but in today's world it seems a bit awkward when weighing it against the possibility of making a compilation mistake. (Okay, I use a Makefile for my compiles, but I still think it's something the compiler should really be doing itself.)
For anything beyond basic writing (which to be fair is often more than enough), the language is full of exceptions and inconsistencies. I've always found that to be quite frustrating. I'm sure that part or all of it might be because I've missed some fundamental points of the language, but I've not yet been able to identify what they actually are.
I do like the concept behind LaTeX. I realise it's a type-setting language, and that puts certain restrictions on it. It's just that when using it, I've often felt like I'm dealing with a language that was built on some dated 1970's programming concepts and restrictions, and has never seriously improved over all this time. Even though it's probably not as powerful, I find it much more convenient (and enjoyable) to just mark up things I write in some basic HTML, which I'm confident will work much more consistently if I try to do anything non-trivial.
I'll have to side with rolfwind on this one. "Let's mis-label it a release, so people will beta-test it for us" is the kind of idea that really disgusts me.
Now I'm not opposed to smaller incremental releases, meaning less features added, and easier to thoroughly test before release. But nevertheless, I expect "stable" to be just that: stable.
You have to understand that while maybe for you "yay, I contributed a bug report to OOo" or "yay, I dug for a week through kernel sources and made my old ISA SCSI board work" may count as fun, for most people it doesn't. In the real world it's more like "fuck, why doesn't this POS print my document right?" Or I can tell you first hand that at work we're not like "yay, it's so cool that we contributed a bug report", but rather "fuck, I'm opening yet another PMR for this POS software. Someone remind me... why are we using this crap anyway?"
What's attractive about OSS to most people is the "because lots of other people have inspected the code and made it better for you" part. It's not the "because you too can spend weeks debugging our code and fixing our bugs, or just beta-testing our unstable stuff and waiting for months for a fix" part. Forcing people to be beta-testers against their will, isn't really going to make your software popular.
A polar bear is a cartesian bear after a coordinate transform.