User talk:Robert Ullmann/SC recovery

Constructive discussion only. Abuse goes elsewhere.

Consensus?

I just I'd note that automated edits have, in my opinion, a greater need for consensus than regular edits. Regardless of what side of this issue an editor might find themselves on, I think it must be admitted that this issue is extremely controversial. Thus, I plead with you, Robert, to wait until a consensus has been reached to do any automated anything in this area. Now, one of the primary points (though certainly not the only one) brought forward by those in favour of a Serbo-Croatian header is the pointless work required to make 3/4 headers with identical sections. If a consensus can be created by using automation to nullify this extra work, then that would be fantastic. While I stand by my position of favoring a unified header, I would be satisfied with any consensus at this point, so that we can all get on with our lives. Just my two cents. -Atelaes λάλει ἐμοί 08:15, 20 August 2009 (UTC)

Quite; the bot is presently set to restore deleted sections (a number of which have content that was elided in combining them into SC-only). I might have spent the last month or so of wikitime on creating some sort of sync-bot, (useful for lots of things), if not having to spend the time dealing with the SC-Über-alles proponents. If you read carefully through all the noise, you'll figure out pretty fast that the "duplication" is not the issue, but just a strawman to push a POV. If the complaint was about the "work", then that is a SMOC, if the complaint was about the duplication in the DB being hard to maintain, ditto.

The bot (at present) is/would be very simply restoring sections improperly deleted. It could easily be viewed as simply reverting vandalism (deletion of content).

Creating/duplicating sections can't be done right now (at least for nouns): when Štambuk created the {{sh-decl-noun}} template he reversed the last two forms from the order for sr and hr; and then in haste to convert entries, changed one or the other to sh-decl-noun without checking, while deleting the valid sections. It can't be fixed by changing the template, as all of the new SC-only entries have the template filled in correctly. So there are now 5,320 entries, all of which must be manually re-checked. Haste makes waste .... the SC sections are full of errors that are not in the standard language sections. Robert Ullmann 08:44, 20 August 2009 (UTC)

{sh-decl-noun} was designed to have the same case ordering as {hr-decl-noun}, in a Croatian grammar tradition, as opposed to Serbian which reverts the last two (as I've been told). This was a concession for the usage of Serbo-Croatian language name, that Dijan and I agreed on. I've checked all the instances of {sh-decl-noun} usage manually, and if there are few instances where the new {sh-decl-noun} has the declenion other that the previous {hr-decl-noun} in the last 2 cases, I'd be grateful if you'd inform me. --Ivan Štambuk 11:17, 20 August 2009 (UTC)

Disagree with a few of the points above, but this is not really the place for such an argument (besides which, I think all the arguments have been beaten to death already, and further reiteration would be pointless). However, I would like to note that a good half the community (myself included) would not see such unification as "improperly deleted" or "vandalism," but rather as "reformatting." So, again, I plead with you not to take any actions unless consensus is clearly documented. -Atelaes λάλει ἐμοί 09:00, 20 August 2009 (UTC)

In the report, you'll see sections that are marked with "*"; they (often) contain content that was removed while "re-formatting". (Inconvenient differences, for the most part ;-) Audio has been dropped, language examples and citations. (in one case very old Croatian quotes, predating even the 19th c idea of SC ;-) And while you are quite correct that it has been beaten, apparently not to death. (;-) You do understand that calling the actual effect of deleting someone's language "re-formatting" is really offensive, don't you (yourself)? Some others may not. (What the hell difference should it make? Well, in fact it does make a profound difference. And it is of course odious.)

Without restoring the sections deleted automatically, it will take editors a lot of time to manually restore the standard languages. Which they will do, sooner or later (a few already have been). So do we do it automatically or manually? Robert Ullmann 09:15, 20 August 2009 (UTC)

90% of of pre-19th cenutury "Croatian" writers did not left a single written evidence of themselves calling "Croats" or "Serbs". The only reason why they are called today "Croatian writers" today is because the concept of nation, when it was created in the 19th century, was retroactively projected onto the ancestors of people how now think of themselves as Croats. The only real "ethnic" designation that was used until the C19 was some general supraregional such as "Slavic" or "Illyrian", together with Croat and Serb, and this latter pair had much, much less usage than the first two. Today Croatian neo-historians try rather pathetically to "prove" that Croat = Slavic = Illyrian, but there is a mountain of evidence that suggests that these term were not synonymous.

Again and again: how exactly is "offensive" ? Croatian language as we know of it today was primarily fabricated in the 1990s. The language that appellation refers to is exactly the same language denoted by the appellations such as "Bosnian language", "Serbian language" or "Montenegrin language", i.e. the Neoštokavian dialect spoken by 21 million people , whose regional divergence in standards is less then that of English in New York and London. While some of the nationalists might be "offended" by that undisputable fact, or even try to imagine that Serbian and Croatian are as divergent as Italian and French (as some of them claimed), or that they are native speakers of "Croatian only" (as e.g. Elephantus (talk • contribs) claims), these statements are nothing but lies.--Ivan Štambuk 11:13, 20 August 2009 (UTC)

This bot is pathetic

The tactics that Robert Ullmann tries to do is the following:

1. Invite Croatian nationalist such as Elephantus (talk • contribs) that will come and waste countless hours simply copy/pasting existing ==Serbo-Croatian== entries to ==Croatian==, and not to ==Bosnian== and ==Serbian== (as if the 15 million non-Croats speaking the same language don't use those words - which Elephantus very well knows they do, it's just that he cannot see beyond his Croatian nationalist nose)
2. Force Dijan and I to maintain ethnic balance be recopying the copied sections to ==Bosnian== and ==Serbian==, again wasting time in a process that requires no human intelligence at all.
3. Robert Ullman afterward comes as a "savior" with this bot, that can bot-automate steps 1 and 2, and restore the so-called "standard languages". --Ivan Štambuk 11:23, 20 August 2009 (UTC)

More petulant personal abuse from Štambuk. I apologize to everyone else. (And if it is just "recopying", that "requires no human intelligence at all", then it would be perfect work for a bot, eh? But of course it isn't. And of course he isn't "forced" to do anything at all; it's a wiki, people contribute what and where they want.) Robert Ullmann 12:11, 20 August 2009 (UTC)

The point is that it shouldn't be done in the first place. It's a complete waste of time for contributors, and even more a waste of time for those who'd want to actually want to learn something by looking at 5 sections, which are (almost) identical, wasting their time figuring out where the trivial differences occur (if they occur).

The only reason why this copy/pasting works now is because somebody actually took time to create fairly complete entries. It does not scale with maintenance and gradual-improvement editing which is 99% of all wiki-activity. Elephantus is acting like a bot, and that's why his behavior can be automated. Not because people "usually" act like bots while contributing SC entries.

Of course people do whatever they want, but we have 6 new Sebo-Croatian contributors in the last 3 days simply because you canvassed them, to do the dirty "prove different languages" work. Which they cannot do, even if they added million neo-Croatian neologisms. As long as the paradigms of the verbs "to be", "to have", "to give", numerals 1-100, pronouns, prepositions and 10 000 basic words inherited from Proto-Slavic are the same - they're all the same language. --Ivan Štambuk 13:28, 20 August 2009 (UTC)