5 ms·
Considering the inclusion of Pandas in the book, I doubt manager types and econ guys are the target group. IMO, engineers would be the ones interested in applyi
by Jolter 5y ago
Considering the inclusion of Pandas in the book, I doubt manager types and econ guys are the target group. IMO, engineers would be the ones interested in applying Python to Excel work (because they realize how ridiculous VBScript is), and they are pretty likely to have install permissions on their machine.
- kingkongjaffa 5y agoIn non tech companies it's super rare to have install permissions in my experience (Manufacturing/Automotive/Aerospace etc.) most companies are running proprietary software from a few key engineering software players, and very rarely there's some team writing legacy but hyper specialised FORTRAN / C / C++ engineering analysis tools. It's a big issue to build tools for others to use - an excel file on a shared drive somewhere might be developed by one person with python installed and elevated permissions but used by 50 with no python installation etc. IT arn't going to adjust their policies for a small group unless they have significant organizational sway.
- carlosf 5y agoInstalling python does not require admin permissions in Windows, just choose $HOME/... as install path. It also does not require admin permissions in Linux, although most distros do not offer a happy path for that.
- kingkongjaffa 5y agocorporate IT in the companies I am talking about is significantly more locked down, python is not install-able by going to https://www.python.org/ https://www.python.org/ and downloading the binary there. Downloading executables is locked down. Not really sure how this is a rebuttal, maybe what you say is true on a default windows installation, not on an IT-dept managed system.
- kitrose 5y agoYou can install Anaconda without admin.
- Jolter 5y agoI’m sure there are corporate users who would like to use this tool but won’t be able to install i. However, I do not think there is any call for posting dismissive comments solely based on that. It does not detract from the value of the tool, for those who can get it.
- kingkongjaffa 5y agoHow is my comment dismissive? I'm sharing my experience of this exact problem we have right now. Users cannot install python on their machines, in an org of 60k engineers, asking for changes to a small group of users comes with heavy resistance from IT. On these machines (most computers in most offices in every western organization) no software can be installed from online, and running any form of executable requires IT/Admin elevation. It's like you guys have never worked in an office in the real world that isn't dev/tech.
- patmorgan23 5y agoI think your vastly overstating how locked down most companies workstations are. Sure if you work for a fortune 500 they might be but there are 10,000 other companies where that probably isn't the case.
- CleaveIt2Beaver 5y agoReally? I work for a <50 employee manufacturer/distributor, and we're pretty firmly locked down. If people want to _update Java_, they need an admin password.
- andrewlgood 5y agoI wouldn't sell the financial analysts short. Pandas was developed by financial analysts to assist in their routine analytics work.
- sterlinm 5y agoYeah, there is a lot of ridiculous usage of Excel in corporate America, but the idea that econ/finance/MBA types are not aware of Python/pandas these days is a farfetched. There are tons of resources like this. https://quantecon.org/ https://quantecon.org/
- deregulateMed 5y agoI wrote Random Forest in VBA. Programming is what you make of it. That said, VBA is slow.
- ptx 5y agoIs it slower than Python, though? It seems intuitively like it should be faster for a lot of things, since it has non-boxed numeric types, early binding, etc.
- deregulateMed 5y agoI'm not sure, selecting data from cells was slow. There might be ways to get around it. I did all sorts of optimization, but looking back on it, I probably shouldn't have saved anything in cells.