[ArchEc] Repec: archive pdf

[Thomas Krichel]

  Guillaume.HORNY at banque-france.fr writes

> As a return, we would be fine with the Foundation's logo appearing
> on the webpage of the Archec project and the other types of support
> refereed to into the application (links to works in which the
> Foundation is involved

  This is done. 

> and advertising through NEP reporting emails).

  There I will need to know exactly what you want to do. At the moment
  all advertising opportunities are free in the sense, I have not sold
  anything. I wanted to try selling from in the second half of the
  year, provided, as I expected, that I loose my Russian job. Anyway,
  there are two types: issue sponsorship and report sponsorship. Issue
  sponsorship is for an issue, i.e. on all reports with a certain
  data. Report sponsorship is more static, it would appear on the
  report for many issues. The report sponsorship has some rather
  stringent formatting rules that come from the way the report issues
  are formatted when they are displayed in text.  Issue sponsorship is
  a bit freer. Since I am coming late to the reply, we can extend the
  period for you to place adverts. The easiest would be for you to do
  a report sponsorship for nep-mon, the text of which we could then
  change to incorporate some message that you want to place. We could
  run this advertising until the end of 2020.

  I feel ashamed to come back to you on that part of your message so
  late. When I got the award, I was ill, I got more ill until about
  middle of April. By then a ton of undone work had accumulating,
  including two complete server migrations, with sponsors continuing
  to pay for two server as the new server was not ready. The timing of
  the migration was somewhat beyond my control, it fell badly.  Thus
  does not excuse that I procrastinated over this.

  At the danger of rambling on, I expected my Russian small job to end
  in June, but it did not. I now more firmly think it will end by the
  end of the year. It pays $800 a months, roughly half my current
  income so it's important money from me. I spent too much time on it,
  running the risk that working on it would endanger me finishing the
  100 hours for Losheim. At this time, I have registered 20 hours work
  done. This is coding only. I don't count time spent on sysadmin at
  all, and don't count research on how to solve a coding problem, and
  neither do I include time spent on examining data. Like today, I
  spent time looking at 32 warc files where the headers say it's PDF
  but can't find PDF in the payload, to determine what is wrong. I now
  have about 29 cases left of various sorts of data that one can make
  a PDF file for. I plan to store them in a particular "bedlam" folder
  and provide for ways for an indexer to bring them back to life, once
  I, or somebody else, can get them converted into PDF.

  Anyway the forecast is that is less than 15 hours more, I will have
  a basic downloader that will take proper account of multiple
  full-text links and multiple historic payloads. Then the task is to
  make some crude optimisation to run in more often with recent data,
  and then I can start on the tricky part, i.e., the integration of
  historic data. Now 20 hours may seem very little and yes, it is too
  little for now, but it is good to underline that a lot of sysadmin
  prep work had to be done. It's still time spent but not counted.
  
  Again, I profusely apologise my procrastination. The good thing
  is that I am procrastinating on the project itself, I feel
  good about what I have done. It's pretty much free of spaghetti
  code.
  
-- 

  Cheers,

  Thomas Krichel                  http://openlib.org/home/krichel
                                              skype:thomaskrichel