131–140 of 193 posts
Excel bashing is easy but, where's a truly comparable alternative? - Alteryx at $5000/year/user? - A database that requires specialized skill or going through a DBA for every little need? - Salesforce? It's hell to configure if you don't have the skill. - Sharepoint? Again. If you're a small business or nonprofit, you've gotta hire or train someone before this is a serious conversation. I would in no way suggest that…
Excel is a great case study. Turns out, if you give people an accessible and widely available Touring complete environment, they will manage to do pretty much anything with it. On another note, it's interesting to observe that the problems spreadsheets have are the same problems that plague poorly designed software everywhere - lack of documentation, lack of testing, lack of input validation, among others.
This is pretty much the same reason that PHP is much maligned - it certainly has gotchas but they're not so much dangerous than, say, the C++ gotchas... When everyone is able to "have a go" at a tool they'll all abuse it in their special way to get things done and good on them for doing it since if every data problem required an engineering team to solve everyone's business would grind to a halt.
I dislike PHP just as much as the next guy, but I am quite impressed with how Hack turned out. It's Facebook's improved PHP. See eg https://hacklang.org/
It's not that Hack is a good language in absolute terms. But I think it's about the best language they could have come up with when starting from PHP.
(For a better comparison instead of C++, but still in similar spirit to C++, perhaps take D.)
They key here is not about swapping tools, or swapping people, because tools and people are both equally fallible. The key is "Code Review" and reasonably proof by testing, not just unit testing which is just nice, but integration testing which is strong. Responsibility is knowing the scale of potential impact, that's when you should 100% insist having some proof with a proper tester.
Excel is a great 'programming language' to quickly _write_ business applications. Especially for relative lay-people.
But Excel is almost impossible to _read_. And it's very hard to make changes, or track them sensibly.
Earlier quoted context omitted.
Knime. Desktop app is free, is not as polished as alteryx but with some community love it will improve massively. It already is WAY better than Excel. Stuff is organized in a manner someone unfamiliar with data handling or programming can understand it. It takes a little bit to master, an some stuff (like looks and variable passing) is still unintuitive.
Is it just a better "Excel" or does it have some actual feature to prevent people making mistakes?
It uses a visual data flow approach to transform data step by step from one form to another. This approach is much better for data transformation. It is also less error prone as operations take place on columns or rows of data, not cells. But you wouldn't try to create a set of company accounts with it.
You’d be amazed at the types of errors we’ve found over the years. Formulas referencing blank cells, relative references formulas that were meant to be static when the formula was copy and pasted to other cells. These were in some pretty serious models (e.g. trading models, risk monitoring models).
One point on Excel, it’s wildly powerful. We’ve done things in Excel that would impress most of this group. Things like 3D formulas (formulas that go through tabs), array functions (which are very infrequently used but are very powerful). I’d argue that most pieces of code (with the exception of recursion and other advanced things) can be replicated in Excel.
Half of the time when validating complex models implemented outside of Excel, we’d use Excel to replicate the logic to benchmark the result.
My big question to this group is, does this type of practice exist in non-banks? Do other industries do this?
Having consulted in large enterprises, the main reason spreadsheets exist: A knowledge worker needs some form of automation support to do the job. Choices are: (a) spend a day drafting a rough spreadsheet, then over the next months tinker with it making it a bit better, more efficient and add more relevant functionalities. (b) put in a formal project request to IT. This will be put into the backlog, and most probably…
Earlier quoted context omitted.
Excel being the de facto standard, Sheets has to try to be compatible with it, quirks and all.
Speaking of quirks, Excel still incorrectly assumes the year 1900 is a leap year. Even Excel has to try to be compatible with Excel. https://docs.microsoft.com/en-us/office/troubleshoot/excel/w...
Excel is a triumph of software. Never before has there been a REPL so powerful, a programming language so widely used, an environment so simple even mere mortals can produce magic - without it what would we have? The power of software would remain closed behind the Ivory gates of the self-titled engineers with development cycles, sprints and countless hours to make a single formulae change. Fear the excel less world,…
> without it what would we have? In an ideal world, we would have pgsql (the PostgreSQL front-end) and AWK, and people who realize that thinking about the model before you start playing with layouts and copy-pasting formulas is a good idea. Edit: Excel is a Dunning-Kruger syndrome enabler. People think they can do data analysis because they can type "=sum(B1:B4)" into a spreadsheet.
Earlier quoted context omitted.
Not just Excel; Google Sheets too. Our school system has student codes in the format ABC0001 where ABC is the first three letters of the surname, and the number is sequential. So a student with a surname Martin might get a code MAR0004. Which Sheets, in it's wisdom, decides to interpret as the date March 2004. Go figure.
Dates can be weird just about anywhere. I recently hit a bug where all data was being checked to see if it was a datetime, and python interpreted `st1` as September 1st of the current year. https://github.com/mozilla/frost/pull/330
Excel is a triumph of software. Never before has there been a REPL so powerful, a programming language so widely used, an environment so simple even mere mortals can produce magic - without it what would we have? The power of software would remain closed behind the Ivory gates of the self-titled engineers with development cycles, sprints and countless hours to make a single formulae change. Fear the excel less world,…
> without it what would we have? In an ideal world, we would have pgsql (the PostgreSQL front-end) and AWK, and people who realize that thinking about the model before you start playing with layouts and copy-pasting formulas is a good idea. Edit: Excel is a Dunning-Kruger syndrome enabler. People think they can do data analysis because they can type "=sum(B1:B4)" into a spreadsheet.