CommitStrip

This seems like a long version of my favourite commitstrip: https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre…

Using advanced devops tools for IoT is always interesting. I have seen a few demos that were variations of the following comic : https://www.commitstrip.com/en/2016/05/26/the-internet-of-things-a-rev… One demo using Microsoft Azure IoT hub, docker, digital twins (a glorified json from memory), and a raspberry pi was fun because it took minutes to deploy something to make a light blink.

I'm the co-founder of Budibase, an open-source low code platform, and 70% of our signups (30,000) are from IT. Low code abstracts a lot of the coding effort, but it's still inherently complex to build applications. Repo for ref: https://github.com/Budibase/budibase Coding is only one part of the puzzle. For us, it's more about speeding up the dev process for IT profs and removing the repetitive, grunt work. Low code, against what some people may think, does have its place - CRUD apps, customer portals, internal tools. But let's not get ahead of ourselves. Low code platform creators are doing the industry damage by promising their tools can provide the world and more. This is not the way to earn trust or respect, and thankfully Budibase does not follow this path. Fun comic strip that is sort of relevant: https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre…

Great comic, which has been going for a long time now: http://www.commitstrip.com/en/

Drag 'n Drop have always had shortcomings, as for AI... I think this CommitStrip sums it up: [0] The punchline for those that don't click: > Do you know the industry term for a project specification that is comprehensive and precise enough to generate a program? Code. It's called code. Programming may change, though I doubt that will be coming in the near future, but it will always exist. Programming is the art of change a specification into a form that can be best read by a computation device. [0] http://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-prec…

Sadly there are two definitions of server floating around these days. There is the classical "box in the corner sending and receiving data over the network". And then there is the new "bunch of software in a container in the cloud somewhere". And with the latter definition, stable and reliable seems to be of lesser importance in the eyes of the faithful. This because they can just fire off a new instance of the container if the existing one falls over (Uptime by machinegun as i like to think of it). http://www.commitstrip.com/en/2015/07/08/true-story-fixing-a-self-ddos…

> You have to get over yourself about the tools not being elegant https://www.commitstrip.com/en/2016/07/11/a-bad-workman-blames-his-too… Jokes aside, the history of mathematics is a pretty worthwhile subject to study to truly appreciate the importance and implications of tools (or mathematical discoveries in this case). There's a reason why it took humans 2000 years after Euclid first wrote his 'Elements' to discover non-euclidean geometry...

Programmers are susceptible to venality just like anybody else. http://www.commitstrip.com/en/2015/09/22/software-worth-billions/

Maybe? To the extent that that could work, I would imagine that I, personally, would be happy reading the textual description instead of watching the video, and for me, we'd now be even closer to text wins 100% of the time. In other words, it's not that you _can't_ give excellent descriptions that would obviate the need for video, it's just that people _don't_, even, or perhaps even especially, when they think they do. If someone writes text that creates a video that shows exactly how to get something apart, then _presumably_ they also watch the video to make sure it works. So the video becomes a debugging tool for their instructions. Perhaps not as good as watching 100 people do it, but maybe even better in some ways. So the video codec you describe could be a useful tool to help create more programmers. https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre…

- The ability and want to learn. I offer the following advice to every firefighter I teach: the day you think or believe you know everything, is the day you quit. (I'm a software dev. by trade with fifteen-odd years experience as an operational firefighter (and still am to this day)). - Find a domain or two you have an interest (better yet, a passion) in, that isn't IT, and pair your IT skills to it. My day job is (more generally) emergency management, so I pair my geospatial analysis and programming skills to solve problems within the emergency management domain. I don't think AI will completely take over, simply because of [1], though it will somewhat remove or reduce the need for 'generalists', where much of the workload is handled by AI, at the prompting of a software developer or engineer, much like how many farm hands were replaced by farming machines when they became a thing. That said, I think we're still half a century off AI truly replacing most software developers. Humans will still be required to 'know what 'good' looks like' and ensure that whatever slop AI spits out actually fixes the problem. This is where domain-specific knowledge is incredibly valuable. See how you can apply your skills to make your interest or passion field better, using AI to your advantage, by exploiting your technical skills with domain-specific knowledge. Most people think of firefighters as just that - putting out fires. That's only partly true though. We use a shiny, and fast, red…

Awesome! Now that I've seen this framework, it tempts me though... I have this personal side project which in the wee days of me career I wrote in CherryPy; I love Python and have been exploring frameworks. I didn't really get a lot written in it because eventually I realized the project is frontend-heavy and I was never in the mood to write the JS for it. Recently I had the need for this side project again so I thought, why not rewrite it using a framework that would avoid me having to deal with the madness that is the Node.js ecosystem? For one reason or another I ended up with Vaadin. I've nothing against Java. If it works, it works. I got over the initial bump that is Spring and I got a "core" backend running, with tests too. My grievance with it is the usual Java fault of needing to write a lot of boilerplate just to shoehorn my app into the framework's philosophy. (Man, part of my initial consideration was even the tooling, editor, autocomplete, and all that, because damn me if I try Spring without these modern conveniences!) And this...this is all the things that attracted me to Vaadin---but in Python! I am so tempted to rewrite again. My experience and your documentation should make this a two hour job, right? Right? ("Haha, I'm in danger," my side project chuckles like Ralph Wiggum, staring blankly at https://www.commitstrip.com/en/2014/11/25/west-side-project-story/)

Those who made such "simulation" probably do have forgot or never learn a classic engineering practice (at least in Europe engineering) witch is named resilience. We need resilient systems as more resilient as bigger is the impact of their eventual failure. In other terms stating we can made a big and complex enough smart grid to run "almost" an renewable means spend an incredible amount of resource to build a hyper-fragile unmaintainable monster that can malfunction at best most of the time. Also the "little storage" they state is not little at all and most importantly it does NOT LAST LONGER. So far we have some long-lasting storage (pumped hydro) witch is very effective where you have enough mountains and basins, witch might be true for let's say Norway and to a little extent Swiss, but certainly not for let's say Germany. Compressed air storage seems to offer an option at a sufficient scale, but so far only some experimental plants exists and they have a significant amount of fragility and risks. Long story short: we haven't enough storage on scale to deploy almost only intermittent sources of power. BEVs can compensate an unstable grid for homes, but not for industry, hospitals, big infrastructures etc. Long story short those who claim such possibility follow this classic: https://www.commitstrip.com/wp-content/uploads/2020/05/Strip-Appli-sto… sorry I do not have one in English but I think anyone can translate quick enough. A video equivalent is…

TLDR: Computer program UX design has to settle for boring consistency and stop reinventing the UX wheel. Learn the best UX practices for your platform based on ACTUAL USER STUDIES and apply them. UX design is like yeast in bread. Too much, the bread overflows the pan and burns everywhere. Not enough, the bread does nothing. Maybe the 90s UX designers didn't use enough and thus we had boring UIs. But today we have 99% yeast and 1% flour and oh it is ugly. Hang the fads, no more gesture-first uninuitive magic PLEASE! Styling is not evil, contrary to the 'unstyled HTML only' crowd, but consistency first and foremost makes for good UX. Consistent (accessible, localizable) look & (undo-able, default cancelling data-destructive) behavior above all, then look at 'intuitiveness' etc. My parents were on the cutting edge of technology. My dad had self-taught and setup auto-dialling dial-up WiFi accessed over PCMCIA cards that he and my mother used for desktop emails, eBay when it was brand new, 320x240 video messages, even some skype. There was nothing they couldn't do on the computer with a little bit of googling and poking around the Control Panel or registry. Today, not so much. And little wonder. User-hostile OS, application design abounds. All of the geniuses at these companies have cargo culted stupid stuff to a grotesque degree: - flat button icons instead of 3d buttons with text (it saves localization costs!!!! at the cost of user confusion) - hamburger menus instead of…

Excel remains a tool. It can be a really powerful one I agree. It's simple to use, a lot of people knows how to handle it. And it's probably one of the best tools to do some quick data analysis. However, like every tool, it has its pros & cons. One of the main cons, IMHO, is that data tends to be mixed with processing a lot (but it also simplify usage, so...). Overall spreadsheets have been really helpful. Retrospectively, there is a good reason why it triggered the personal computers boom in the 80ies. They are good adhoc tools that can help lot. And Excel is the dominant one. However, there are tons of horror stories around Excel, worst case scenario, it has lead to wrong decisions affecting the world economy. I've personally a few horror stories (fortunately, not affecting the world economy). At my last job, we used an internal spreadsheet to evaluate costs and margins when answering contract bids. This spreadsheet was a monstrosity taking ages to compute and it was a nightmare to manage, sync, and share between all the participants. I've also seen the traditional "on what did you work last week" spreadsheet, quite annoying to update, even more given that my workstation is under Linux... I've also seen spreadsheets as specification documents, generating an xml file, which in turn, through some xslt, was used to generate C++/C++ headers files describing an interface between two sub-systems. The concept (automating spec to code translation) was really interesting.…

I recommend reading 'The Wealth of Nations'. As an economics textbook, it is fairly dated, but it makes the point that ten people with a specific specialization can do more than ten people all doing the same work at once. Therefore the value you can provide increases with specialization as you can be part of a team than can provide more. My salary increased once I marketed myself as having a specific specialization, but the difficulty of finding a job increased too. It is a basic application of supply vs demand. As, say, a PHP developer, you're competing with millions of other PHP developers around the world. There are plenty of jobs, but there are also plenty of people who are competing with you, driving your price downwards. If you narrow it down to knowing a lot about a very specific framework or PHP system - for example, you know a lot about Laravel or Drupal, then you're competing with fewer people, and people are willing to pay more for an expert, but there are also fewer potential jobs. There's also an associated risk. If you specialize in Laravel and Laravel goes out of style, you will have to remarket yourself as a PHP dev again... Some people specialized intensely in Microsoft Silverlight, and they ended up like this - http://www.commitstrip.com/en/2015/07/28/betting-on-the-right-horse/ (it's not a total loss, as some programming paradigms work across languages). With the risk comes increased reward.

Well, customer data isn't stolen by actual hacking, in my experience it's humans. So many companies, particularly younger ones, have zero interest in putting up barriers to access as the company grows because in the early days, everybody was trustworthy and "because bureaucracy bad". So all the customer emails, phones, addresses, birth dates (and, I'm guessing, in the US SSNs) routinely fly around in Excel files called something like "Order Metadata Report" and sent to 50 people in 5 departments each of whom has their own use for it (like counting customers). Judging by the Sony hack it's not just SMEs. If you want to steal data from a company, just pay a student a few hundred bucks to take up an unpaid internship in marketing (particularly anything to do with emails or customer segmentation) and give him a USB key and teach him some VBA and basic SQL (making him useful for reporting). The interns always end up running the reports so have a lot of access, usually complete access - financial information is the only thing that's not shared around. More advanced companies have a shared database access built into the excel files with a single login for everybody which never changes (hello 300 angry users) so with a copy of this file, you have perpetual up to date information long after you're gone. Then you try to stop them from doing this and the C-level folks will say something like "it's OK just this time" and "please stop slowing us down". Most of them will be gone to the…

I don't see how you think that's a cop out. Instructions are never laid out perfectly: https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre… Holding LLMs to the standards of human contextual understanding without communicating sufficient context can be dispensed with, as fantasy. Using LLMs you quickly learn how much can be inferred from existing, which is dynamic and particular to each model today. There will always be a gap in how much instruction is needed due to mismatches of existing versus intent. Current state, you don't need much to get a lot out that can be verified prior to merge.

> Instead of writing lines of code you are writing requirements. https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre…

> We have yet to design a language to cover that, and it might be just a donquijotism we're all diving into. We have a very comprehensive and precise spec for that [0]. If you don't want to hop through the certificate warning, here's the transcript: - Some day, we won't even need coders any more. We'll be able to just write the specification and the program will write itself. - Oh wow, you're right! We'll be able to write a comprehensive and precise spec and bam, we won't need programmers any more. - Exactly - And do you know the industry term for a project specification that is comprehensive and precise enough to generate a program? - Uh... no... - Code, it's called code. [0]: https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre…

Still relevant . . . https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre…

I'll just put this here as a reminder: https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre…

>sufficiently descriptive natural language specification https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre…

No one can write machine code from memory other by writing machine for years and just memorizing them. Just like you can't start writing Python without prior knowledge. > Analogously, I imagine future software dev to consist mostly of writing specs in natural language. https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre…

My opinion on this whole "using LLMs as a programming language" thing is described nicely by this comic: https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre… > Do you know the industry term for a project specification that is comprehensive and precise enough to generate a program? > Code, it's called code.

People who are terminally capitalist have been salivating over this idea basically since this hype curve first started. Someone made a solid joke about it as far back as 2016: https://www.commitstrip.com/2016/08/25/a-very-comprehensive-and-precis…

"A very comprehensive and precise spec": https://www.commitstrip.com/en/2016/08/25/a-very-comprehensive-and-pre… Switch in "prompt" in stead of "code", if you will; still pretty much same thing.

CommitStrip has an awesome comic about this exact process. https://www.commitstrip.com/en/2014/11/25/west-side-project-story/? Shameless plug, I made a YouTube video that points out a bit of nuance that the comic doesn't explicitly point out. https://www.youtube.com/watch?v=-L37U81MunM