I strongly prefer to have draft quality periodic reports/memos to be read by the team members than a tons of crappy Jira issues gathering dust and spreading confusion.
189 karma · joined September 27, 2018
I strongly prefer to have draft quality periodic reports/memos to be read by the team members than a tons of crappy Jira issues gathering dust and spreading confusion.
The problem: irreproducible results.
Once you chain dozens of command lines, unversioned shitty scripts, no md5 checksum-ed inputs/intermediate results etc. you are in a smelly swamp full of monsters. Even you one week later will not be able to do it again.
Markdown docs/jupyter notebooks are minimum even if one does quick and dirty proof of concept thingy IMHO.
I got a data sets probably not suitable for loading into a memory table (close to 1000M rows CSV). I did split it into 20M rows chunks, read one by one into a DuckDB temporary table and exported as parquet.
SELECT using glob prefix.*.parquet where mycolumb=foobar does work but can be a bit faster. Apart from sorting the input to parquet CSVs, what can he done? The CSV chunks were already sorted.
python -m nuitka --clang --follow-imports main.py
repo: https://github.com/MRCIEU/gwas2vcf
If someone can make the program run faster by whatever means, it will make a bunch of people quite happy.
In short: shit on input does not bide well
Blood-brain barrier is another obstacle to cross. Quinidine taken with loperamide blocks transport protein P-glycoprotein and causing loperamide to have opioid like effects in the brain.
If a drug is metabolized in the liver then instead of ingesting a bottle of pills (most likely not safe, certainly not recommended), the way to bypass the liver is to use suppositories. Again, safety-wise it is not a sensible thing to do.
In the old not so good days a driver could encounter at night a horse driven buggy with no lights or reflective elements whatsoever. Just few days ago at night while going downhill on a steep road I have seen a well camouflaged hiker walking on my side of the road and stepping aside luckily in time.
I will take reflective vest on a biker/pedestrian/hell, even on a horse anytime I am driving/biking and can hit them after spotting them too late.
Did the total number of the road accidents stayed the same in the whole period?
Number of kliometers driven per year?
Number of the cyclists on the road?
Therefore Markdown should have higher priority than some crappy level Word or Excel skills. Even if 80% of endusers will use Markdown wiki/docs just as viewers, there is no point of hogtie sysdmins/developers and ask them to document in Word, google docs and keep data in excel files
On a separate note: one can script simple link checking/verification of the fetched content. And flag all pages with dead links, or with links where PDF/html/whatever did change from a snapshot taken say every month or so.
Instead of a forced page renewal there should be a bonus for fixing highly useful pages and a bounty for finding errors in the documentation.
Otherwise the same crappy page with few nuggets of useful links/info will be copied verbatim for ever and ever.
Servers: Debian stable or Centos 8 (depending on the job place)
Also Mx Linux where I need Debian/Ubuntu-like system with a pleasant GUI.
Looking into Clear Linux and NixOS.
Few features (apart from be able to view wiki diffs) from my point of view:
1. Wiki edits in activity
2. User wiki edit history (= edits times, pages, size of diffs)
Maybe in a long run it will be good to have Markdown based collaborative document editing. Meaning that each editor can branch the draft doc, introduce changes. Then these branches will need to be reviewed and merged, probably by 1-2 people.
Hope it helps