[Bioc-devel] Binary file (flash SWF) in source package?

Martin Morgan mtmorgan at fhcrc.org
Tue Oct 6 16:46:08 CEST 2009


Gábor Csárdi wrote:
> Martin,
> 
> thanks a lot for your answer.
> 
> About the issues you mentioned. The binary is a single file and it is
> 400Kb. It is (well, supposed to be) completely platform independent.

The package guidelines

http://wiki.fhcrc.org/bioc/Package_Guidelines

suggest that the source package resulting from running R CMD build
should occupy less than 2MB on disk, so you are likely ok on this.

Martin

> As you mentioned, licensing should be OK. (Contrary to the license of
> Flex itself, that has some interesting things. But luckily, these do
> not apply to the binary created with Flex.)
> 
> Best Regards,
> Gabor
> 
> On Tue, Oct 6, 2009 at 4:05 PM, Martin Morgan <mtmorgan at fhcrc.org> wrote:
>> Hi Gabor --
>>
>> Gábor Csárdi wrote:
>>> Dear BioC developers,
>>>
>>> what is your current position regarding binary files in BioConductor
>>> source packages? I have a package that visualizes overlapping
>>> biclusters, and the interactive visualization part in written in Adobe
>>> Flex. Flex creates a simple platform-independent SWF binary file that
>>> can be used with most web browsers.
>>>
>>> Putting this file into the source package has the advantage that one
>>> does not have to have Adobe Flex installed to install the R package.
>>> In addition to the binary, the Flex source code would be also included
>>> in the package, of course, and the package license would be GPL.
>>> Adobe Flex is free, as free beer, but it is not free or open source
>>> software.
>> Thinking out loud a little...
>>
>> Some of the issues are licensing (ok in this case), added security
>> responsibilities associated with binary file redistribution,
>> cross-platform compatibility of redistributed binaries, and potentially
>> substantially increased package size.
>>
>> On the other hand having unusual system dependencies has proven to be a
>> significant challenge for us, and for our users -- even 'simple'
>> installations can be complicated when there are multiple build machines,
>> there can be challenging version-matching problems, it can be difficult
>> to convey the system dependencies to the end users, etc.
>>
>> In this particular case I think we can proceed, but this isn't meant to
>> be a general endorsement.
>>
>> Martin
>>
>>> Thank you, Best Regards,
>>> Gabor
>>>
>>
>> --
>> Martin Morgan
>> Computational Biology / Fred Hutchinson Cancer Research Center
>> 1100 Fairview Ave. N.
>> PO Box 19024 Seattle, WA 98109
>>
>> Location: Arnold Building M1 B861
>> Phone: (206) 667-2793
>>
> 
> 
> 


-- 
Martin Morgan
Computational Biology / Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N.
PO Box 19024 Seattle, WA 98109

Location: Arnold Building M1 B861
Phone: (206) 667-2793



More information about the Bioc-devel mailing list