Paul Cantrell
I’ve been around a while too, and I can vouch for this.
I remember somebody telling me with deadly earnestness circa 1999 that Java Beans were going to make it so that business analysts could write integration code themselves, no developers. Java Beans.
Two certainties in software:
1. The work always changes.
2. The work never goes away. https://front-end.social/@alvaromontoro/114494970685095188
Paul Cantrell
The argument with Java Beans was that you can just drag and drop connections between the getters and the setters, so what’s left to code?
Yes, really.
Paul Cantrell
People keep making the same mistake, again and again and again and again forever, of thinking that it is syntax that makes software development hard.
Oh honey.
Re this from @mathaetaes:
https://infosec.exchange/@mathaetaes/114656764053846137
(P.S. Visual coding is actually really cool, and IMO an underexplored PL design space — but is very much coding, and very much tricky for the same reasons as any other kind of coding.)
Javier
@mathaetaes @inthehands let's not forget Drools, where people can define business rules for the system directly in a spreadsheet, Java imports and all. No need for developers there.
Paul Cantrell
Wow, I had in fact forgotten that! If it’s in a spreadsheet, it’s not programming, of course.
Mathaetaes
@inthehands lets not forget the “visual coding” hype, where people were going to generate all their code straight out of a UML model. No need for developers there.
Paul Cantrell
@fishidwardrobe https://hachyderm.io/@inthehands/114422202675741454
Fish Id Wardrobe
@inthehands the real trouble comes when staff actually fall for this and write code.
I've spent much of my career coding in one of these, and I've met people with no understanding of coding at all who have written thousands of lines of garbage which kind of works – often by copying and pasting the same dreadful code again and again.
Paul Cantrell
@tomjennings
Max/MSP is one of the langs I have students in my P-Lang class research present to the class. “Convince us it’s a language!” is the mandate, and it’s always a fun day in class.
I know a fair number of experimental musicians who are not at all programmers, but got into Max/MSP and did some wonderful things with it. It’s quite marvelous as a tinkering tool for individual artists! Has a bit of the HyperCard spirit I praised elsewhere in my posts.
And yes, it doesn’t scale up well. “Write-only” is a bit severe IMO, but it certainly does not scale well to large projects or multiple people.
tom jennings
Lots of art folk adopted Max/MSP as a non programmers language. It's nicely graphical, connecting inputs to outputs with lines, etc.
It's virtually uneditable. It's another write -only language.
As faculty we'd get students who genuinely wanted to learn to code but in a two or three year program could take maybe 2 semesters total.
People skilled at Final Cut Pro, quite complex and sophisticated software, thought programming was like operating FCP; just tell me where the controls are and I'll be fine.....
I tried to get across that like many things coding is a skill, requires dedication and hours and mistakes.
Very few ever put in the effort. It wasn't entirely their fault and they weren't lazy. The whole story is tl;dr of course.
cliffordheath
@inthehands As I speak in these definitive terms, I have recently revised a semantic formalisation of Object Role Modelling, a graphical formal language. We formalised it by mapping each graphical feature to an abstract syntax, and formalising that. But it's hard work, and still unproven in terms of use of differencing in engineering review
Paul Cantrell
@cliffordheath
Yeah, version control and also editing are big barriers to graphical languages. My assertion is just that it’s an underexplored research space — like electric cars were in the early 20th century.
cliffordheath
@inthehands syntax is what makes software engineering possible. Linear texts are amenable to differencing through LCS, hence to version management, on which all SW eng is built. Graphical languages cannot(*) be differentially managed, so however cool they are, they cannot be engineered.
LCS is the fundamental algorithm of software engineering.
GNU social JP is a social network, courtesy of GNU social JP管理人. It runs on GNU social, version 2.0.2-dev, available under the GNU Affero General Public License.
All GNU social JP content and data are available under the Creative Commons Attribution 3.0 license.