jerven

44 ms·

421.
by jerven 11y ago
Tasting the chefs cooking it much better than dogfood
422.
by jerven 11y ago
The program of the talks can be seen here 2015.biohackathon.org/symposium
423.
Biohackathon symposium 2015(ustream.tv)
4 points by jerven 11y ago|1 comments
424.
by jerven 11y ago
This is really neet! I will share this with the rest of the biohackers at http://2015.biohackathon.org/hackathon as a nice approach.
425.
by jerven 11y ago
Much worse, overdrafts are a mean product! Especially in the UK variety. When I was an intern in the UK I once handed over a cheque a day before I got the cash. The cheque bounced, I was forced to pay 25 GBP in overdraft fees and the person
426.
by jerven 11y ago
Three years one of the DBA's at Credit Suisse mentioned that their DB for their most important business line DB was 150 TB in size. Can't remember anymore if that was an oracle installation, or Sybase... So this is gossip level in
427.
by jerven 11y ago
While the title is "Giant bags of mostly water", the subtitle is better for understanding what you are going to read. An interesting quote is "The most unprotected system on your network is your sysadmin's workstation.&
428.
Securing your IT infrastructure by securing your team [pdf](kernsec.org)
2 points by jerven 11y ago|1 comments
429.
by jerven 11y ago
Pharma is in that space. They don't advertise it but they love "alternative" medicine because profit margins are insane. 20 euros for 10 grams of sugar brilliant, and it doesn't even have to work :) What is not to love
430.
by jerven 11y ago
Its more about data interchange support, i.e. we could support GraphML instead/next to the RDF varieties. But this would be difficult for us to generate in a streaming way. Then for our end users, we would need to hack in a namespace c
431.
by jerven 11y ago
4-5 was a very long time ago in graph db time :) neo4j and its competitors have changed the lot!! While neo4j has it's proponents. The lack of standards support means that as a data provider it's hard to support.
432.
by jerven 11y ago
Yes and it has gotten much better in recent releases. On the other hand switching to an other sparql impl takes about 2 weeks of work in evaluation and deployment. We can currently switch to GraphDB or oracle sem net without further effort.
433.
by jerven 11y ago
I think the wikidata was the right decision :) SPARQL really is lovely when dealing with public databases over http. But then I am biased by the success of sparql.uniprot.org :) On the other hand the usecase of wikidata and uniprot are quit
434.
by jerven 11y ago
You turn this capability off. I did and I sleep much better since!
435.
by jerven 11y ago
It does help this use case, as you can restrict which role's a user/connection can have. i.e. you can't set role to anything you want if you are not a superuser.
436.
by jerven 11y ago
In the developing world fuel logistics are a major cost. Local solar+battery is quickly cheaper in the long run in areas where infrastructure is lacking. However, access to capital is a major challenge. i.e. the upfront costs means the chea
437.
by jerven 11y ago
At this point peerj will be fine. Peer reviewed is what matters at this point. Also find a Prof whose work you like and are interested in. See if you can use some of their published work to do something neat.
438.
by jerven 11y ago
It called "embarrassingly parallel" because it is embarrassing for the Cray and other supercomputer vendors sales people when their expensive hardware does not outperform a 5 year old solution that cost 1/2 their fancy gear ;
439.
by jerven 11y ago
That should be Piz Daint at CSCS, so Super conductor behavior, weather and many other things. Mostly non embarrassingly parallel problems, where its high speed interconnect pays off.
440.
by jerven 11y ago
Yes and I completely agree, but a slow AKKA seems easier than rewriting if from scratch in C#. e.g. seemed a smaller investment to me.
441.
by jerven 11y ago
Instead of re implementing could something like ikvm work for their users/usecases?
442.
by jerven 11y ago
Your data is more important than your code. Is the often neglected fact in bioinformatics. Whatever you do document your file formats.
443.
by jerven 11y ago
I don't think that is fair 1.5 introduced the enum keyword. That was not a fun change in a codebase I worked on. Java 6 and later are more sticklers in what is allowed by the api contacts and not breaking those than in the sun days.
444.
by jerven 11y ago
There a number of JVM's written in Java with an mini interpreter in machine code. All code starts in the interpreter. GC and JIT run on themselves, more and more code gets optimized and no longer runs in the mini interpreter but actua
445.
by jerven 11y ago
The real problem is that the nuclear power plant will easily have 5 orders of magnitude construction costs associated with it. And historically a 10 times longer construction period. i.e. you have 7 years no power then a 1 GW is added to th
446.
by jerven 11y ago
Those are basically unpaid posts
447.
by jerven 11y ago
SPARQL is used this way at eli lily. All databases are more or less accessible via SPARQL using e.g. d2r or any other conversion tool. Technical performance is crap, yet the all important financial performance is fantastic. Basically all th
448.
by jerven 11y ago
Most likely there is legal advice available for you for very little money. Ok, I am not in the US, but in Switserland you can have an 30 minutes of lawyer time for 50/100chf via local services. Similar things exist in the Netherlands.
449.
by jerven 11y ago
Its of topic but I think it was a great idea to have terrible C interop. It really forced people to write java all the way. This meant the JVMs could really evolve unlike the Python and Ruby ones. I am not sure the new java FFI is such a gr
450.
by jerven 11y ago
Interesting presentation showing java to assembly and back again. When can the optimiser kick in and when does it need to back off. Code samples showing behaviour in the matching github https://github.com/dougqh/jvm-mec

More ›