| Time |
S |
Nick |
Message |
| 03:29 |
|
|
hydrajump joined #sourcefu |
| 03:32 |
|
|
aditsu joined #sourcefu |
| 04:14 |
|
|
aditsu joined #sourcefu |
| 07:29 |
|
|
sivoais joined #sourcefu |
| 08:27 |
|
|
tumdedum joined #sourcefu |
| 13:11 |
|
|
hydrajump joined #sourcefu |
| 15:56 |
|
|
aditsu joined #sourcefu |
| 22:17 |
|
sivoais |
I came across this today: <http://beautifulopen.com/> "Beautiful Open: Beautiful sites for Open Source projects." |
| 22:19 |
|
pdurbin |
sivoais: neat! shared! https://botbot.me/freenode/opensourcedesign/2015-03-25/?msg=35053969&page=2 |
| 22:20 |
|
pdurbin |
sivoais: we're indexing 303,826 documents into Solr: http://irclog.perlgeek.de/crimsonfu/2015-03-25#i_10340663 |
| 22:20 |
|
* sivoais |
joins YAIC (IRC channel) |
| 22:20 |
|
pdurbin |
heh |
| 22:22 |
|
sivoais |
ooh, nice! I shall have to play with that to compare it with ElasticSearch |
| 22:23 |
|
sivoais |
a friend of mine worked on improving a Grails Solr plugin <https://github.com/zeebo404/grails-solr-plugin> |
| 22:24 |
|
pdurbin |
cool |
| 22:24 |
|
pdurbin |
now I'm trying to break that job up. lots of Solr docs. takes a while |
| 22:25 |
|
pdurbin |
the idea is to spread the work across multiple servers |
| 22:26 |
|
pdurbin |
partition the work into 3 parts or 4 parts and give each app/web server some of the work of talking to Solr, of indexing into solr |
| 22:27 |
|
pdurbin |
I'm not sure what the standard terms are for this though. |
| 22:27 |
|
pdurbin |
partitions? |
| 22:27 |
|
pdurbin |
slices of work? |
| 22:29 |
|
pdurbin |
westmaas: what would you call it? |
| 22:30 |
|
pdurbin |
prologic: ^^ |
| 22:30 |
|
sivoais |
I'd use partition, shard, or node |
| 22:31 |
|
pdurbin |
ok so 3 paritions |
| 22:31 |
|
pdurbin |
3 partitions* |
| 22:31 |
|
pdurbin |
and what do you call each one? I need to give them IDs |
| 22:32 |
|
pdurbin |
partitionId=1? |
| 22:32 |
|
sivoais |
that seems reasonable |
| 22:33 |
|
pdurbin |
good, 'cause that's what I'm going with |
| 22:33 |
|
sivoais |
I'd read through posts on <http://highscalability.com/> to see what the common terms are :-P |
| 22:33 |
|
pdurbin |
hmm. ok |
| 22:34 |
|
pdurbin |
curl 'http://localhost:8080/api/admin/index?numPartitions=3&partitionIdToProcess=2' {"status":"OK","data":{"dvObjectIds to process":[2,5,8],"numPartitions":3,"selectedPartitionId":2,"availablePartitionIds":[0,1,2]}} |
| 22:34 |
|
pdurbin |
that's my interface for now. hope it makes sense to people |
| 22:34 |
|
pdurbin |
hope it makes sense to me months from now :) |
| 22:35 |
|
sivoais |
hahah, yeah, that's what documentation is for! :-) |
| 22:36 |
|
* pdurbin |
writes moar docs |
| 22:36 |
|
sivoais |
I'm so glad when I have to read the docs for my own code. I'd be so lost without them. :-P |
| 22:36 |
|
sivoais |
"I wrote this, but I don't know how to use it now..." |
| 22:37 |
|
pdurbin |
I've been writing more unit tests. |
| 22:37 |
|
pdurbin |
they're almost better than docs sometimes |
| 22:37 |
|
pdurbin |
which can get out of date |
| 22:39 |
|
sivoais |
Depends on who the target audience is. If I was writing something like "ls" or printf(), looking at tests won't help me. They'd be too fine-grained. |
| 22:40 |
|
pdurbin |
oh sure |
| 22:43 |
|
prologic |
generally if I want to reimplement something |
| 22:43 |
|
prologic |
I observe it's behavior and interactions with other systems |
| 22:44 |
|
prologic |
a spec can help though :) |
| 22:44 |
|
pdurbin |
prologic: hi! so! partitions? did you see what I was saying about splitting work across nodes? |
| 22:45 |
|
prologic |
lol |
| 22:45 |
|
prologic |
was that your blog article? |
| 22:45 |
|
prologic |
wait |
| 22:45 |
|
prologic |
waht |
| 22:45 |
|
prologic |
oh |
| 22:45 |
|
prologic |
hang on lemme read |
| 22:46 |
|
prologic |
yeah it's called partitioning up the data or work into slices/partitions :) |
| 22:46 |
|
prologic |
I wouldn't use node here |
| 22:46 |
|
pdurbin |
ok, I was thinking maybe slices as well |
| 22:46 |
|
prologic |
that's more in the context of a node that does part of the work :) |
| 22:46 |
|
pdurbin |
but I think I'm going to stick with parititions and partitionIds |
| 22:47 |
|
prologic |
*nods* |
| 22:47 |
|
pdurbin |
cool |
| 22:47 |
|
pdurbin |
I'm doing a little modular math. It's fun! https://github.com/IQSS/dataverse/commit/c4fb067 |
| 22:47 |
|
prologic |
partitions/slices are synonyms anyway :) |
| 22:48 |
|
prologic |
oh man I wish I could read Java better :) |
| 22:48 |
|
prologic |
haha |
| 22:48 |
|
prologic |
Atlassian just contacted me -- but the conversation ended quickly when I said I hadn't done Java since uni :) |
| 22:49 |
|
prologic |
like what are you mean to do really |
| 22:49 |
|
prologic |
stack hacking on Java projects in your free time just because? :) |
| 22:50 |
|
sivoais |
hehe...Java has changed quite a lot in recent years |
| 22:50 |
|
prologic |
it certainly looks that way |
| 22:50 |
|
prologic |
it seems to have borrowed the decorator syntax from Python |
| 22:50 |
|
prologic |
which is not surprising it's a very powerful construct and nice way of expressing it |
| 22:50 |
|
prologic |
but honestly I just can't bring myself to really go back to things like Java or C# |
| 22:50 |
|
prologic |
C is tolerable :) |
| 22:51 |
|
prologic |
*just* |
| 22:51 |
|
prologic |
:) |
| 22:51 |
|
prologic |
no offense pdurbin :) Java is a fine language, paltform and ecosystem |
| 22:51 |
|
prologic |
*cough* |
| 22:51 |
|
prologic |
if you're used to it of course :) |
| 22:52 |
|
prologic |
I wish I could find that article on the web that talked about the no. of things a programmer had to know about a language |
| 22:52 |
|
prologic |
it compared many popular languages including Io |
| 22:52 |
|
pdurbin |
I just work here. |
| 22:52 |
|
pdurbin |
:) |
| 22:52 |
|
prologic |
Io of course was on the bottom of the no. of things you had to know at ~6 I think |
| 22:52 |
|
prologic |
Python is about ~23 IIRC |
| 22:52 |
|
prologic |
haha |
| 22:52 |
|
prologic |
if anyone finds that article linkk it here |
| 22:52 |
|
prologic |
it's very enlightening |
| 22:53 |
|
sivoais |
actually, Python borrowed Java's syntax (Java calls them annotations) <https://www.python.org/dev/peps/pep-0318/#background> |
| 22:56 |
|
sivoais |
It would be pretty cool to have a map of where different syntax constructs originated |
| 23:04 |
|
sivoais |
prologic: Io is hard to search for. :-P |
| 23:04 |
|
prologic |
o'really? |
| 23:04 |
|
prologic |
haha |
| 23:04 |
|
prologic |
There is some history in Java using @ initially as a marker in Javadoc comments [23] and later in Java 1.5 for annotations [10] |
| 23:04 |
|
prologic |
I did not realize this :) |
| 23:04 |
|
prologic |
haha |
| 23:04 |
|
prologic |
sivoais: Io is nigh impossible to searhc for |
| 23:04 |
|
prologic |
I agree :) |
| 23:05 |
|
sivoais |
you would think that designers of new programming languages would try to choose more searchable names... |
| 23:06 |
|
sivoais |
but we now have things like C, R, D, Dart... |
| 23:07 |
|
* sivoais |
names the next big programming language "Computer". |
| 23:10 |
|
prologic |
lol |
| 23:11 |
|
prologic |
I have mio-lang :) |
| 23:11 |
|
prologic |
but it's still a toy haha |
| 23:13 |
|
sivoais |
traits++ |
| 23:15 |
|
prologic |
heh |
| 23:56 |
|
|
aditsu_ joined #sourcefu |
