Young Hands Club

August 19, 2019

Week 6 Plan

Filed under: Shreyas Ragavan — Shreyas Ragavan @ 3:55 p.m.
  1. [ ] v.py : 10 lines, minimum.
    1. My intention is do more than 10 lines, but I will set 'smaller' targets untill I gain confidence.
  2. [ ] Complete partially finished Python basics revision.
  3. [ ] GLMnet results issue - spillover from Week 5.
  4. [ ] Hash out project plan for exploring Open Data, and using it as practice. Datasets of 'interest'
    1. Computer price index
      1. Computer price index - by type of purchaser, monthly link
    2. Wages
      1. Wages, salaries, employers social contributions. by province and territory link
      2. Employee wages by occupation, annual link
      3. Employee wages by industry, annual link
      4. Wages - 2017, 2018 link
  5. [ ] Learning about software and the software industry
    1. [ ] Summarise / Discuss: http://ossasepia.com/2017/02/18/the-open-sores-of-monkey-code/
    2. [ ] Summarise / Discuss http://ossasepia.com/2017/06/12/o-brave-new-code/
    3. [ ] Read: http://ossasepia.com/2017/12/21/eucrypt-correcting-mpi-implementation/
      1. [ ] Understanding this article will need me to first complete http://www.dianacoman.com/2017/12/07/introducing-eucrypt/, which has been suggested earlier. I did start reading the same, but wonder if it would be better to attack this after gaining some traction with V.
    4. Summarise / Discuss: http://trilema.com/2019/no-platforms/
  6. [ ] Improve summary of 'Your feelings are to get you', based on Diana's comment. (1 paragraph).
  7. [ ] Learn + Summarise :- systemd | init ( Refer log)
    1. Guidance on articles to definitely read would be helpful.
  8. [ ] Formulate response to Diana's comment on 'what is data science'. Worth noting that this was discussed before. My 'view' has improved, since then, but I'm not sure I would answer very differently.

2 Comments »

  1. 8. The problem is that what you gave earlier is not much of a definition especially if you claim the whole thing to be a "science". Hopefully the problem with this situation is clear. Maybe scope it better/choose something clearer so that you can actually make sure you are working towards it.

    7. I wouldn't really spend time on this right now. It's just *one* example of many and if you don't know about systemd you are probably better off just avoiding it entirely. Perhaps leave it as lowest priority, if there's any time remaining.

    6. OK.

    5. The MPI-correction is more about an unexpected bug uncovered in code reused ("open source") than about EuCrypt as such. Hence this particular aspect you should be able to get even reading the article on its own. Yes, it is part of the EuCrypt series and yes, there are *some* parts that you won't get unless going through the hole EuCrypt but those can be left for later. The point there is to see a concrete example of the multi-faceted trouble with "open source" as a "piles of code by anyone for nobody in particular". As such, I think it stands on its own quite well.

    4. Ok, let's see that.

    3. Ok, but what's the priority here?

    2. Either it's complete or not, lol. Is there a plan for this? How exactly do you know (as in: can really say, with concrete arguments) that it's "complete"?

    1. Ok. Just actually include in the write-up a clear discussion of the lines in question and otherwise ask in #ossasepia when/if stuck - Stan (asciilifeform) wrote the thing after all and he offered already to help you if you have questions, you know?

    Comment by Diana Coman — August 19, 2019 @ 5:43 p.m.

  2. (8) >> Maybe scope it better/choose something clearer so that you can actually make sure you are working towards it.

    Yes. I think this vague definition has been at least 1 significant impediment behind progress (related to the small steps directive for v.py). (4) and (3) are actually the specific tasks related to (8).

    (7) Okay.

    (5) Right. I will do it.

    (3) : nobody is waiting on me to deliver this. Low(er) priority - but considering that I do know something of the theory already (and it's unanswered/forgotten on the forum) - it will be significantly faster to tackle than the earlier (similar and also pending) UMAP task from Week 2. All of which was delayed due to the foray into Docker. If things go well - (3) should be done today (without docker).

    (2) : LOL... I should bookmark "Complete partially finished". I wanted to convey that I did something and not so much remains. It is actually a 'course',(https://www.dunderdata.com/offers/7EaLyimi/checkout). I'm done about halfway and somewhat familiar with the rest, and it was more a prep for v.py (1) and a general revision.

    (1) : I'm on it.

    Comment by Shreyas Ragavan — August 20, 2019 @ 4:34 p.m.

RSS feed for comments on this post. TrackBack URL

Leave a comment

Work on what matters, so you matter too.