unhammer

44 ms·

1. 451.▲
   Higher-Kinded Data (types for both validated and raw data witho redundancy)(reasonablypolymorphic.com)

   2 points by unhammer 9y ago|0 comments
2. 452.▲
   by unhammer 9y ago

   Quill has a great API, it seems like they really thought things through and designed it (what's a good word for the opposite of ad-hoc?). Even the lower-level bits are available with a clean API if you should ever need them (e.g. it&
3. 453.▲
   by unhammer 9y ago

   I agree SpaCy's doc's are great, though I still managed to miss the point about the small models missing real word vectors, which http://mlreference.com/word-vectors-spacy notes at the very top. I wasted some time
4. 454.▲
   by unhammer 9y ago

   Huh, first time I've heard of Mozilla Lockbox ( https://www.bleepingcomputer.com/news/software/firefox-to-ge... / https://mozilla-lockbox.github.io/lockbox-extension/ ). That's
5. 455.▲
   by unhammer 9y ago

   I've been using the ColorHug 1, and found it worked quite nicely. The included Fedora CD (yes, this was some time ago) let me calibrate just fine. Once I wanted to do it straight from XCFE, I ran into a few challenges, but it wasn&#x27
6. 456.▲
   by unhammer 9y ago

   `bind` is usually written as `>>=` in Haskell. Haskell also lets you write simply print instead of \d -> print d I wouldn't consider this part syntactic sugar – if one did, I suppose one would have to say th
7. 457.▲
   Flight Sim Company Embeds Malware to Steal Pirates’ Passwords(torrentfreak.com)

   403 points by unhammer 9y ago|169 comments
8. 458.▲
   by unhammer 9y ago

   https://www.youtube.com/watch?v=ojZbFIQSdl8&feature=youtu.be... is a nice explanation for curly-brace programmers of why it's called a sum type (vs structs, which are product types)
9. 459.▲
   by unhammer 9y ago

   I've also noticed the thing on mobile Firefox. It's not unusable, but there is something weird going on.
10. 460.▲
    by unhammer 9y ago

    ugh, my line breaks got lost. Trying again since I can't edit: * Projects ** Customer X *** TODO issue 42 breakage SCHEDULED: <2018-02-12> ** Customer Y *** Fooproject **** TODO fix slow frobni
11. 461.▲
    by unhammer 9y ago

    I keep all paid work in a work.org file and clock in/out of headings there. The basic structure is * Projects Customer X * TODO issue 42 breakage SCHEDULED: <2018-02-12> Customer Y * subproject foo TODO fix slo
12. 462.▲
    by unhammer 9y ago

    OTOH, I feel it's taken too far in the regex packages. They feel extremely daunting for someone just getting started with Haskell. E.g. the wiki says that Text.Regex.TDFA is the thing to use. First try: λ> "abc" =~ &
13. 463.▲
    VISL CG-3 Is the New Assembler (Constraint Grammar Is Turing Complete)(wenkokke.github.io)

    2 points by unhammer 9y ago|0 comments
14. 464.▲
    by unhammer 9y ago

    does it relate in any way to Isenkram? http://people.skolelinux.org/pere/blog/Isenkram__Appstream_a...
15. 465.▲
    by unhammer 9y ago

    How's performance on that, have you compared with plain Windows or wine?
16. 466.▲
    RIO: A standard library for Haskell(github.com)

    1 points by unhammer 9y ago|0 comments
17. 467.▲
    by unhammer 9y ago

    See https://blog.jle.im/entry/practical-dependent-types-in-haske... where you get this function definition: https://github.com/mstksg/inCode/blob/master/code-samples/de... I li
18. 468.▲
    by unhammer 9y ago

    related: http://libregraphicsworld.org/blog/entry/embroidery-design-o...
19. 469.▲
    by unhammer 9y ago

    magit ftw: https://i.imgur.com/a8Mvbkf.png Though I'm sure Emacs has loads of other vulnerabilities :)
20. 470.▲
    GIMP 2.9.8 Released, with On-Canvas Gradient Editing, Wayland screenhooting(gimp.org)

    4 points by unhammer 9y ago|0 comments
21. 471.▲
    by unhammer 9y ago

    > JMAP will enable various long-desired features (e.g. snooze, delayed/undo send). says chrismorgan at https://news.ycombinator.com/item?id=15856942
22. 472.▲
    by unhammer 9y ago

    > (there are now frogs in my office when people come visit me, they get "woken up" and I have to explain to them) How do you explain to your frogs that you woke them up?
23. 473.▲
    by unhammer 9y ago

    There is a possible defense: https://www.dwheeler.com/trusting-trust/
24. 474.▲
    GSM ciphering indicator in Android(issuetracker.google.com)

    1 points by unhammer 9y ago|0 comments
25. 475.▲
    by unhammer 9y ago

    > I don't really see what the big game changer is whether a laptop comes preinstalled with Ubuntu or if I do it myself. It means a certain level of support. They're actively trying to make this thing run Linux well, instead of
26. 476.▲
    by unhammer 9y ago

    I've been a happy customer of https://www.fastmail.com/ for some years now (and used Gmail before). I highly recommend them for anyone considering leaving Gmail. You pay them with money instead of data.
27. 477.▲
    A bug-fix 12 years in the making: Windows Unicode Support in OCaml 4.06.0(dra27.uk)

    85 points by unhammer 9y ago|13 comments
28. 478.▲
    by unhammer 9y ago

    This is pretty cool! But I would've liked to see some eval on smaller datasets (or just a graph of BLEU vs training set size), since the main use for something like this is where you don't have parallel data, in which case the mon
29. 479.▲
    by unhammer 9y ago

    https://en.wikipedia.org/wiki/Interlingual_machine_translati... is an old concept, used to be seen as The Right Way (as opposed to the pragmatic way) of doing MT back in the day, before the success of (very pragmatic,
30. 480.▲
    by unhammer 9y ago

    off-topic, but that's a pretty cool e-mail address she's got :-) I wonder how many clients fail on it.