Discussioni template:PageQuality

Da Wikisource.
Jump to navigation Jump to search

Sorry to write in English. I think that this is important.

Before you decide to use the same categories in the 'pagina' as in the 'main' namespace, I suggest that you take time to understand the implications of proofreading text when scans are available. It does not make sense to blindly apply the same categories and definitions.

For example, on it.ws, SAL25% means "no source" (fonte non indicata) or "incomplete".

However, in the 'page' namespace, there is always a source : the scan is the source. Therefore "fonte non indicata" is not relevant.

In addition, a single page is very unlikely to be "incomplete", because it is much shorter than a whole book. In general, if the text of a page is incomplete, it means that there has been a problem transcribing it (for example that the scan is corrupted). This kind of problem desserves a special category, and the level 2 is made for that.

However, "no source" or "incomplete" should not be the default status of a page.

ThomasV (disc.) 21:34, 23 giu 2008 (CEST)

Sorry my friend, but I don't agree at all. I understand that you'd prefer that we use the same logic defined in other Sources, but our SAL system is good also for Page namespace. Look here. Exclude the first column (the source one) and you'll understand how it works here:
  1. Pagine SAL 25% --> Page not completed or empty
  2. Pagine SAL 50% --> Page completed but not rigltly formatted. This is the situation that you call "Level 2". A text completed but wiht some problems
  3. Pagine SAL 75% --> Page completed and formatted
  4. Pagine SAL 100% --> Same as 75% but with anoter one user that check it
As you can see we don't need to use anoter one approach since our fits really well and it is yet known by all users.
We are adapting your new system right now, so don't worry that we'll prepare a good help page explaining the levels in a good way. So please wait to see the definitive one.
Just to be clear, I like all the work you've done and you're doing to make Wikisource a better place but I personally don't like this approach. I don't understand why we should use fr.source/en.source logic. We don't need an intermediate level because we have yet 50%. And I don't like that other communities decide for us. We have been using SAL system for a long time, it would be nice to contact us just to listen to our ideas. As I told you, it can be really frustrating to look inside pages/code/templates to know the new things. You are really nice and your suggestions helped us to adapt the system (right now Indice:Critone looks in a really nice and coherent way) but try to understand that all these new things came to us with no discussion and this makes it all much more difficult.
But this is not a problem. We can use your system anyway in a really good way. We have different category names but the result is still the same. The only limit, a tecnical one, is that at the end of edi pages we see a strage list of levels (2-1-3-4) but, knowing that depends on things that we cannot change, we can go on. -- iPorkscrivimi 22:34, 23 giu 2008 (CEST)
I Just slipped in here to reassure Thomas about iPork's views. In this particular project the definition "incomplete or without source" is plainly understood: if a page is a scanned source itself SAL25% will simply mean that its transcription is incomplete. We already have texts currentli uploaded and not transcribed, and one of our goals is to have such pages transcibed by users (a simple task: see Wikisource:Aiuta Wikisource). So SAL 25% can mean "You can see the source, but it's not yet transcribed.
In the Page namespace formatting has a more radical meaning than in "online copypaste": you can check de visu if formatting can be considered right or wrong, complete or unacceptable. If a page is transcribed and well formatted is almost done, namely SAL 75%
If there's a problem about SAL 25% applied to scanned pages it can be disambiguated on the help page. What I feel important is a link to help pages and a plain description on them, that is what we are striving to do. I want to point out that iPork is stating policies decided through long discussions, whichs explains our need to adapt a "great but a little stranger system" to our policies. Let's go on and if there's any doubt, let's talk it over. - --εΔω 23:31, 23 giu 2008 (CEST) (CEST)
I am sorry if you feel that things come without discussion. The ProofreadPage extension has never really been used on it.ws, and I hope that the reason is not people here having that feeling. As a developer, I listen to the advice of the people who use my software, not of those who do not use it. I suppose that you can understand why.
For the moment, this extension is still new for you, and I guess that you need to acquire more experience with it. Once you become familiar with that system, I suppose that you will realize that it is useless to indicate that a page is empty: if a page is empty, it should simply be deleted.
I hope that this extension will have some success on it.ws.
ThomasV (disc.) 23:54, 23 giu 2008 (CEST)


Dear Thomas, just my two cents on this topic. For us is really important to talk with you, so we can learn from each other a different approach: I think this is a right and good thing.
I'm trying to figure out your point of view, and maybe the difference is only on a more or less user-friendly approach:
On it.ws the SAL has been discussed and accepted by the community, and there's no need to change it, in front of a small advantage. For us it is important to say if a text is complete, formatted or it needs some wikification or transcription. I know that for your experience this is not so important.
I guess that you think about some cases in which the book image were not so good, or ununderstandable. But actually I don't know if these (rare, I hope) cases could justify a deep change in the SAL structure, that I natuaral and clear for the whole community.
On it.ws we haven't really use the Proofread extension mainly for our fault or different priorities, and we really appreciate your work, for real. It would be nice for us to fit your software in our structure and policies, that are strict and complex but they work out, we can assure. It would be great if you could help us to adapt the extension at our needs: anyway just talk over, it will be useful for sure!
PS: just to tell you no lies, I have a bit of experience of your great extension ;-) --Aubrey McFato 00:52, 24 giu 2008 (CEST)
Scriverò in italiano dal momento che mi risulta più semplice. Già l'anno scorso abbiamo iniziato ad usare l'estensione e chi frequenta it.source ben si ricorda di una iniziativa dedicata per spingere i lettori a lavorare su Indice:Critone.
AIUTA WIKISOURCE: un piccolo sforzo per un grande aiuto

Vuoi aiutare Wikisource ma non hai tanto tempo a disposizione?
Non conosci bene il linguaggio wiki?
Ti piace leggere ma hai paura a modificare le pagine?

Collaborare con Wikisource non è mai stato così semplice! Scegli un testo tra quelli che hanno la versione originale a fronte, leggi una pagina e trascrivi ciò che vedi nell'immagine nello spazio apposito. Ti basterà conoscere solo 3 codici per formattare. Ma non preoccuparti, c'è una facile guida che ti segue passo dopo passo.
Con poca fatica dai un prezioso contributo alla biblioteca.

Vuoi saperne di più???
Era il 29 luglio del 2007 e ti garantisco che tutti qui abbiamo visto nell'estensione uno strumento finalmente semplice e potente. Dopo alcuni mesi ci siamo accorti che era cambiato tutto ed oggi siamo qui a capire come funzionano le novità. Se ben ti ricordi abbiamo anche sperimentato il famoso OCR automatico, ma anche quel sistema da un giorno all'altro non funzionava più. E cosa dovevamo fare? Ci siamo fermati e ci siamo concentrati su altri lavori su source.
Sul fatto che una pagina vuota vada cancellata mi pare una cosa bizzarra, per non dire altro. Io scannerizzo un testo e carico le immagini su commons. Passa Aubrey e crea la pagina Indice. Poi passa Orbilius e inizia a trascrivere alcune pagine. A meno che il sistema non sia così intelligente da creare da solo le pagine attraverso l'OCR nel momento in cui vengono inserite nel namespace Indice, è sempre necessario il livello base con la pagina vuota. Questo per me è Wikisource: un progetto collaborativo aperto in cui ognuno fornisce il suo contributo. Le pagine vuote e senza testo sono alla base del percorso collaborativo. Anzi, credo proprio che il SAL 25% trovi la sua più corretta applicazione proprio sul namespace Pagina. Ora, se in altre Source cancellate le pagine vuote è una libera scelta del singolo progetto. Permetti che qui si sperimentino approcci diversi.
PS: Se poi, usando il sistema capiremo che il nostro approccio è errato almeno ci saremo arrivati per conto nostro, esattamente come abbiamo dovuto fare fino ad oggi mancando del tutto una traccia, una guida, un pezzo di pagina in cui si spiegano queste meravigliose innovazioni tecniche. -- iPorkscrivimi 10:14, 24 giu 2008 (CEST)

Proposta di integrazione tra estensione e SAL[modifica]

Proposta mutuata dalla discussione sopra, in pratica trascrivo quanto proposto da iPork.

  • 2 -> Da trascrivere -> 25%
  • 1 -> Incompleta o non formattata -> 50%
  • 3 -> Trascritta -> 75%
  • 4 -> Riletta (da 1 utente diverso) -> 101%

Che ne dite? --Accurimbono (disc) 12:31, 1 ago 2008 (CEST)