[Bioc-devel] GEOquery for Bioc devel

[Tim Triche, Jr.]

GEOmetadb had an issue with Dropbox jerking around URLs (cf https://github.com/ttriche/simpleSampleSearch/blob/master/R/checkForUpdates.R ); I wonder if this might be related?  I'd hate for GEO to become as flaky as biomaRt.

I should have submitted simpleSampleSearch for 3.3 but it seemed too trivial in its current form. But my judgment sucks (as you all well know).

--t


STOP READING AT THIS LINE
------------------------------
IT'S A HUGE WASTE OF TIME!


Pointless rant:

Every time I hear a new proposal to build an site to "aggregate and share and crowdsource" and blah blah blah, I wonder if any of the folks involved have ever tried to support a useful piece of infrastructure.  "Thankless" doesn't even begin to describe it.  The NAR database issue showcases this kind of zombie project.  Almost all such proposals should instead be gauged as competitors for scarce NIH funding needed by bedrock infrastructure (GEO, SRA, etc). Then again, looking at the sad state of physical infrastructure in the US, I don't know how much that would help :-/. 

At least in Europe (cf. ENA, a godsend vs. SRA) one sees investments in infrastructure on a sensible scale. There are probably many things horribly wrong with the EU system, but as far as I can tell, failure to maintain infrastructure is not one of them.  Perhaps it is down to the aforementioned cultural differences (above)?

Not sure what can be done; perhaps someone like Atul Butte could put this into a "translational" context easily.

/rant

--t

> On May 8, 2016, at 8:46 AM, Vincent Carey <stvjc at channing.harvard.edu> wrote:
> 
> Hmm.  Very important piece of software so we should try to remedy this.  I
> had assigned some GEOquery problems in the edX class and there were some
> people who could not get getGEO to work, but the problem seemed
> intermittent.  There was a workaround posted on the support site which
> involved separate steps for assay and platform data acquisition; ultimately
> the solution data object was just made available directly.
> 
> Can/should we make the build/distribution process robust to data server
> flakiness?  The user may encounter intermittent problems, but the software
> should still be available IMHO.
> Reduce server-dependent tests and examples?
> 
> On Sun, May 8, 2016 at 11:08 AM, Dan Tenenbaum <dtenenba at fredhutch.org>
> wrote:
> 
>> It has not built since the devel (3.4) builds started. Apparently GEO has
>> been pretty flaky.
>> Dan
>> 
>> 
>> ----- Original Message -----
>>> From: "Kasper Daniel Hansen" <kasperdanielhansen at gmail.com>
>>> To: "bioc-devel" <bioc-devel at r-project.org>
>>> Sent: Sunday, May 8, 2016 5:54:59 AM
>>> Subject: [Bioc-devel] GEOquery for Bioc devel
>> 
>>> I don't see GEOquery for Bioc devel.  Is that a mistake somehow?
>>> 
>>>      [[alternative HTML version deleted]]
>>> 
>>> _______________________________________________
>>> Bioc-devel at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>> 
>> _______________________________________________
>> Bioc-devel at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/bioc-devel
> 
>    [[alternative HTML version deleted]]
> 
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel