[ Date Index ][
Thread Index ]
[ <= Previous by date /
thread ]
[ Next by date /
thread => ]
On Wednesday 21 July 2004 20:53, Simon Waters wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 John Daragon wrote: | It may be. But I *still* don't see what this has to do with whether the code | is Open Source or not. Specific instances of a software problem can rarely be traced to whether the code is "open source", or even free software. Since to "the programmer" the source code is generally available (although often proprietary software uses libraries where this isn't the case). Freedom impinges mainly on the process, i.e. how you would address this issue (writing a macro to help underskilled users, versus changes to the spreadsheet itself to avoid them in the first place
This may not be a good thing. If, for example, it breaks the code for thousands of people out there who were using a spreadsheet as ... well ... err ... a spreadsheet. This is a good idea because one user somewhere doesn't know the syntax for a string ? I guess if C had been Open Source, then we may have avoided all those pesky errors where people wrote if (i = 0) when they meant if (0 == i) ? Come to think of it, that's more or less what happened with Java. Every time I look at it it's different. Some programmers think this is cool. They're not usually the ones who have responsibility for the project.
), or what checks and balances apply to the software development and release process. Say if the spreadsheet vendor's developers insert a huge easter egg, or make what would be unacceptable compromises for some users on the algorithmns for which table cells to recalculate, or fail to support your preferred platform, as random <sic> examples. In contrast I suggest Outlook's security problems would not have persisted so long in the free software world, as people who needed more security than the base product offered would have fixed it and made those fixed available to others. There are situations where this process is slow but I think is demonstrable. If there is a general trend it is to free software being more rounded, more interoperable, less buggy.
I've no argument with this at all, although (at least in part because of the inherent limits of its funding model) it has a way to go. For example, I've recently completed my part of an installation of about 1400 workstations for an insurance company. The servers have all migrated from IBM mainframes to Linux boxes, but the desktops are all Win32. The only reason that they're not running RedHat is that OpenOffice couldn't reliably read Word documents. I know that's not OOs *fault*, but other (funded) Word Processors have managed migration in the past. We could have fixed it ourselves, I guess, but we hadn't the time, so the client installed Office.
You expressed concern at free loaders in an earlier post, rather harshly naming Redhat despite the huge development efforts they have funded.
Not me, guv. You're confusing me with Some Other Bloke.
But free loaders aren't a problem generally (although I have an issue with GNU Chess users whose PCs have viruses, where I see a small -ve cost per ~ MS Windows user to myself). But the point is freeloaders cost other users nothing, and they may always discover (and in some cases even fix) an obscure bug before you hit it, or request a useful feature you hadn't thought of.
Actually, I have *nothing* against free use of software at all (I've got 13 RedHat boxes and Mandrake on my laptop). I just think that a lot of it was taxpayer funded. That's not free where I come from.
The more users of a software product in general the better it is for individual users. This only seems to breakdown when the vendor is slow to address issues, which is not a sustainable position, in the free software world this can be addressed by forks or patches, in the proprietary software world this generally results in other types of failure (of both the user and the vendor).
I absolutely agree. The main reason I use Linux, and the main reason I recommend it to clients is that when the vendor fails to fix the problem we have the source. jd (who still thinks that the gene-in-a-spreadsheet thing was User IQ error) -- John Daragon argv[0] limited john@xxxxxxxxxx Lambs Lawn Cottage, Staple Fitzpaine, Taunton TA3 5SL, UK (house) 01460 234537 (office) 01460 234068 (mobile) 07836 576127 (fax) 01460 234069 -- The Mailing List for the Devon & Cornwall LUG Mail majordomo@xxxxxxxxxxxx with "unsubscribe list" in the message body to unsubscribe.