Because those sources vary widely in terms of metadata quality (author, date...etc. records), some of the converted articles look a bit messy. That's the next Hermes version's job: to get a bit wiser in terms of recovering the metadata, and in terms of robustly recovering the internal document references/citations.
Many of the original articles are missing from the converted lot: a few of them were using drawing packages, and Hermes doesn't support drawing yet, a few of them weren't converted because of an unexperienced script which filters the original lot and doesn't do an intelligent renaming of files, and about 5 of them weren't converted because Hermes needs some (small) fixes.
The total number of articles tried is 1005 (751 articles of the month 2005/3, and 254 of 2005/4); the success ratio today is 0.438.
The good news is that this version tries to push/pop parentheses at the table cell border, or around the align marker in (ams)math structures, so that the presentation-MathML result is well-formed with less need of human intervention.
The source distribution is here, and the samples site is here, as usual.
First, in my opinion, this is not open-access, I already explained why. Briefly, this model is, in fact, an author-pays model: it moves the charging of money from readers to writers, so it is doomed to stay against open-access, by its own definition; besides, the U.K.'s Royal Society pointed already the potential bad consequences on research quality of using the author-pays model (an author with results to publish but no money to pay for publishing them is to get lost).
The results of practicing the author-pays model as an option over the last year suggest to O.U.P. that: 'open access is likely to be "only one of a range of models" that will support research communities'.
Secondly, because the author-pays model is not an open-access model, the results of O.U.P.'s publishing experiment have no relevance to open-access.
Paying to be made freely available :). This nonsense (pity for those who are financing it) is due to marketing people in publishing trying to wrap all their reactive moves in phrases containing "open-access".
The research paid partially or totally from public funds has to be made available to the public at no significant supplementary costs: that, in essence, is open-access, and that, in essence, is what the public paid for already.
I will keep repeating this whenever I will get news about various publisher-marketing moves disguised as "open-access" experiments.
Citesc, în lift, lista obișnuită de reguli pe care'o găsește românu' de la bloc:
1. este interzis să...
2. este interzis să...
3. este interzis să...
A rămas un iz feudal în atitudinea asta, fiecare nene cu slujbă are drept de popă peste neaveniți (plătitori):"este interzis să...". Într'un oraș pe bune, ai găsi o altfel de atitudine: "nu băga șurubelnița unde nu'i cazu' pen'că costă y lei". De la comunism la capitalismu' conștient de comunism e o distanță falsă, ca de la biserică la inginerie socială, în atitudinea față de cetățeanu'-tramvai. Ambele îs autorități măsluite, simulacre, doar că ultima pare să plătească tribut cauzalității (are un ce explicativ) așa că'i mai eficace pe termen scurt (cam cât durează viața activă a unui bun platnic).
Pare că românu' prezent încă simuleză bine că'i pregătit s'audă ce'i interzis și ce nu'i, de la oricine se repede să simuleze că anunță distincția. Sara împart o bere amândoi și fac haz de istoria zilei; ar fi și asta o rețetă viabilă, de n'ar plăti doar unu', același, dintre ei.