4 ms·

Also... Excel is great. What exactly is "ahead" of Excel? People on here really underestimate how great of a product Excel is because of its cultural affiliatio

Also... Excel is great. What exactly is "ahead" of Excel? People on here really underestimate how great of a product Excel is because of its cultural affiliations.

t-c-h 6y ago
Excel is great, and generally really useful. But not so much in microbiology. Sure, if you have several dozen genes/proteins with some quantitative values and some conditional formatting to visually distinct certain properties, it's fine. But in microbiology you usually works with very large datasets that undergo a lot of calculations through a lengthy pipeline. Larger sets and more intense ML approaches even need HPC's. Inserting Excel into this pipeline would be catastrophic.
- ethanbond 6y ago
  Oh yeah, I'm totally aware. My day job is designing and building software to construct/maintain exactly those pipelines. What I'm saying (which is not news to you) is that clearly Excel is excelling along some dimension that people really care about. Even if it is lacking along other dimensions, the goal ought to be figuring out what makes Excel so good, then figuring out how to reconcile that with the needs of e.g. a high scale genomics pipeline. But to make even an inch of progress against that, we need to get rid of this "haha Excel is for plebes" attitude that many in the software community adopt.
  - jpindar 6y ago
    Excel excells at being already installed on most computers. That's it.
    - ethanbond 6y ago
      Huh. Most computers also have a built in calculator app, but I have never heard of those being difficult to pull from the hands of users, even when replacing it with purpose built software that costs tens of millions of dollars per year. But what do I know :)
ellis-bell 6y ago
LibreOffice Calc is free and doesn't corrupt data. It also has full support of UTF8
- otherme123 6y ago
  In the case menctioned in the article, LibreOffice is on par with Excel. Don't believe it? Open up LibreOffice, type "MAR-1" or "SEP-2" in any cell, and look how easy is your data corrupted. I have meet some other glitches: identifiers in the form of "1E123" get turned into scientific numbers. The column was something like "1A123", "1B123", etc. Those things are sneaky: you can have thousands of rows, and Excel/LO doesn't mind if only the 0.1% matches its rules for smartness. They just change without notice, leaving the others intact. I'm a bioinformatic, and this kind of stuff is a daily issue. IMO, two problems colide: 1. People with very rudimentary knowledge of computers. I've meet some people way smarter than me, with thousands of papers written, that cannot open a CSV with R with a gun to their head. They can cut you, put three robotic arms in your heart, remove a tumour, sew you and send you home in three days. But R is just too much. 2. People that needs to collect and analyze data, and Excel is the easiest tool they know, but at the same time it's no lame toy: it sorts, it sums, it filters, it makes stats, it graphs... You cannot ask this people to use SQL, specially if it involves foreign keys. They just use Excel for data collection, storage and analysis. Excel (and LibreOffice) are to blame. This shit could be avoided if they had a "scientific mode" to not be smart with your data, and be the default mode for any file in CSV/TSV mode. Unless you explicitly turn the file into a XLS/ODT or ask "turn this range of cells into dates for me" or "this cells are scientific numbers", they should not change a single dot.
  - dragonwriter 6y ago
    > This shit could be avoided if they had a "scientific mode" to not be smart with your data, and be the default mode for any file in CSV/TSV mode. There's a whole lot of business use of CSV/TSV with Excel that benefits from the existing defaults. Heck, I've seen federal government websites distributing code lists that are actually in CSV format with .XLS extensions and that are expected to use the default “smart” conversion. Even if Excel should arguably have had different defaults, the impact on other uses of changing it now would be enormous.

Also... Excel is great. What exactly is "ahead" of Excel? People on here really underestimate how great of a product Excel is because of its cultural affiliations.

t-c-h 6y ago

Excel is great, and generally really useful. But not so much in microbiology. Sure, if you have several dozen genes/proteins with some quantitative values and some conditional formatting to visually distinct certain properties, it's fine. But in microbiology you usually works with very large datasets that undergo a lot of calculations through a lengthy pipeline. Larger sets and more intense ML approaches even need HPC's. Inserting Excel into this pipeline would be catastrophic.

ethanbond 6y ago
Oh yeah, I'm totally aware. My day job is designing and building software to construct/maintain exactly those pipelines. What I'm saying (which is not news to you) is that clearly Excel is excelling along some dimension that people really care about. Even if it is lacking along other dimensions, the goal ought to be figuring out what makes Excel so good, then figuring out how to reconcile that with the needs of e.g. a high scale genomics pipeline. But to make even an inch of progress against that, we need to get rid of this "haha Excel is for plebes" attitude that many in the software community adopt.
- jpindar 6y ago
  Excel excells at being already installed on most computers. That's it.
  - ethanbond 6y ago
    Huh. Most computers also have a built in calculator app, but I have never heard of those being difficult to pull from the hands of users, even when replacing it with purpose built software that costs tens of millions of dollars per year. But what do I know :)

ellis-bell 6y ago

LibreOffice Calc is free and doesn't corrupt data. It also has full support of UTF8

otherme123 6y ago
In the case menctioned in the article, LibreOffice is on par with Excel. Don't believe it? Open up LibreOffice, type "MAR-1" or "SEP-2" in any cell, and look how easy is your data corrupted. I have meet some other glitches: identifiers in the form of "1E123" get turned into scientific numbers. The column was something like "1A123", "1B123", etc. Those things are sneaky: you can have thousands of rows, and Excel/LO doesn't mind if only the 0.1% matches its rules for smartness. They just change without notice, leaving the others intact. I'm a bioinformatic, and this kind of stuff is a daily issue. IMO, two problems colide: 1. People with very rudimentary knowledge of computers. I've meet some people way smarter than me, with thousands of papers written, that cannot open a CSV with R with a gun to their head. They can cut you, put three robotic arms in your heart, remove a tumour, sew you and send you home in three days. But R is just too much. 2. People that needs to collect and analyze data, and Excel is the easiest tool they know, but at the same time it's no lame toy: it sorts, it sums, it filters, it makes stats, it graphs... You cannot ask this people to use SQL, specially if it involves foreign keys. They just use Excel for data collection, storage and analysis. Excel (and LibreOffice) are to blame. This shit could be avoided if they had a "scientific mode" to not be smart with your data, and be the default mode for any file in CSV/TSV mode. Unless you explicitly turn the file into a XLS/ODT or ask "turn this range of cells into dates for me" or "this cells are scientific numbers", they should not change a single dot.
- dragonwriter 6y ago
  > This shit could be avoided if they had a "scientific mode" to not be smart with your data, and be the default mode for any file in CSV/TSV mode. There's a whole lot of business use of CSV/TSV with Excel that benefits from the existing defaults. Heck, I've seen federal government websites distributing code lists that are actually in CSV format with .XLS extensions and that are expected to use the default “smart” conversion. Even if Excel should arguably have had different defaults, the impact on other uses of changing it now would be enormous.