Re: [LUG] Excel mangles genes draft letter

[Carl <cjm@xxxxxxxxxxxxxxxx>]

Adrian Midgley wrote:
> The argument over the model of development of software for scientific use - 
> whether like Gnumeric it should be produced in an open source peer reviewed 
> way, or like Excel it should be secret and proprietary, the correctness of 
> its algorithms garanteed by the reputation of the company selling it will not 
> be resolved today.  In this one instance however, we submit that the Free 
> (libre) software approach has provided a better tool.

It doesn't actually prove a case for open source. It makes absolutely no 
difference to the end user whether or not the source code for Excel is 
available. If you use an application improperly and with insufficient 
knowledge of its functions then it will never do the job you think you 
want it to do.

Excel can be a damn frustrating app in many ways, but it generally 
correctly handles cell formatting. If a user sets up a spreadsheet 
(which is not intended to be used as a replacement for a database) 
without knowing what they are doing (which seems to be the norm, 
particularly in public institutions) or with insufficient skills, then 
they must be prepared to suffer the consequences and not blame the software.

As I said, I don't see how an open source model helps in this kind of 
instance. I also find Gnumeric lacking a lot of the functionality I need 
- I don't see it as a serious contender in the spreadsheet stakes - Open 
Office's spreadsheet is much better and slightly less frustrating than 
Excel in that it handles non-Excel files better.

Carl

--
The Mailing List for the Devon & Cornwall LUG
Mail majordomo@xxxxxxxxxxxx with "unsubscribe list" in the
message body to unsubscribe.