If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Announcement
Collapse
No announcement yet.
The Debate Over GCC's SVN-to-Git Conversion Approach Won't Be Settled This Year
Well there's Maxim's review of ESR's latest conversion on the mailing list, wherein he notes that the resulting repo is missing several branches and basically broken. So the result of ESR's 6+ months of delaying and ego stroking is a tool that's still clearly inadequate.
No one is making new SVN repos, so eventually conversion tools will become obsolete. ESR isn't doing this for the sake of longevity or general purpose tools but simply because his entire raison d'être is ingratiating himself and tricking compiler hackers into believing he's as competent as they are. Unless you've been following him for a long time, it's easy to miss the manifestations of his (by now, well documented) personality disorder.
It's disappointing to know it. I had hopes reposurgeon would be a really useful tool and useful for GCC Subversion to Git conversation too.
Interesting to know. Did he have an appointment with a respectable and competent psychiatrist? He seems to be in a sad situation if all that is happening.
Are there a reliable Maxim vs ESR works comparison? Preferable by a skilled third party.
There is no neutral 3rd party here. Everyone has an agenda. And there is no truly objective definition of "best" history. At this point they are mostly down to arguing over which, and how, email addresses should be extracted, abstracted and possibly modified, which moves into philosophical arguments about how to reference people; some people have had dozens of email addresses over the decades, and some people have changed their names, or their moniker (James is now known as Jim), but it is the same person, maybe. And as anyone who has had a long philosophical debate about how many angels can dance on the head of a pin should know, there is no provably correct answer, even if a number of people seem to really care about the value.
I do wonder, those CVS -> SVN conversion errors were not a problem thus far evidently, why it is suddenly a problem now. Just switch with either method (preferably Maxim's since it's been available for months) and move on.
I doubt that there was ever an actual technical problem here...
Actually, in most cases, the bureaucracy is comprised of workers and managers who have seen many changes of government during their careers, and whom as much as possible ignore the "ruling class"... if they don't like a change of policy, no worries... it'll take years to implement, and by then the policy will have changed again.
Those bureaucrats are real rulers of the world...
Yeah, something like this is happening here.
People should just check how Maxim's conversion works, and use it if it works good. This is a huge shame that they ignore Maxim's work.
I do wonder, those CVS -> SVN conversion errors were not a problem thus far evidently, why it is suddenly a problem now. Just switch with either method (preferably Maxim's since it's been available for months) and move on.
Most of the 1st world governments have more bureaucracy, but those bureaucrats are mostly in place to justify the decision that has already been made by the ruling class (but it is important to pretend to listen to those that want to believe their voices matter).
Actually, in most cases, the bureaucracy is comprised of workers and managers who have seen many changes of government during their careers, and whom as much as possible ignore the "ruling class"... if they don't like a change of policy, no worries... it'll take years to implement, and by then the policy will have changed again.
It seems like using a general use tool for the conversion makes more sense than a ad-hoc set of bash scripts, since the general use tool can allow other projects to benefit from all of the improvements that have gone into it. So reposurgeon seems to be the best choice for this reason, since it can also help other projects.
But it's not really a general-use tool... it's a hyper-specialised tool created for the purposes of the GCC migration, which has nevertheless been implemented in a very abstracted way. Pretty much every other major project, and many big corporations, have already migrated to git — generally using the standard migration tools — so it's likely that reposurgeon will only ever be used for this one conversion project.
Most of the 1st world governments have more bureaucracy, but those bureaucrats are mostly in place to justify the decision that has already been made by the ruling class (but it is important to pretend to listen to those that want to believe their voices matter). Many open source projects do not have anyone (or a very small group) who is willing, able, and trusted to truly lead, willing to make tough decisions, willing to accept the responsibility for pissing off some of the people, and willing to accept being second guessed for years (with future votes and actions trying to undo decisions at every turn). So you get paralysis in decision making. Leadership is not always about being popular. And while it is different, I believe only one member of the Debian Technical Committee that made the systemd decision is still a member, and a number resigned after what was considered a fairly poisonous process to the individuals on the TC (they were, at least, smart enough to know they needed to walk away).
Leave a comment: