In the last 5 years we have blocked access to the internet through bot checks and ad networks. We block screens with ads or consent forms which aren’t even enforceable and make the user bark on command to get past.

We have become trained seals. Captcha popped up? Better bark. Cloudflare challenge appeared? Better bark like the trained seal and click the checkbox.

Nobody even talks about this. The fact these things are annoying is constantly discussed but I have yet to see any article covering the fact we have trained our species to bark when asked without question and what the ramifications are down the road.

There isn’t even a planned offramp from this trajectory, these things are going to get more pervasive and annoying while technology improves. Where doea this actually end?

I think pretty much every dev understands the issue but they are limited in what they can do about it. Quitting a job because they won’t let you optimize is noble but unrealistic for the vast majority of devs.

I would love for optimizations to start being prioritized. More specifically, I would love to see vendors place limits on memory use in apps. For example, Steam could reject any game over 50gb. I do not believe for a moment that any game we currently have needs more than 50gb except maybe an mmo with 20 years of content. Or Microsoft could reject apps that use more than X ram. They won’t ever do that but without an outright rejection, this won’t be fixed.

This is really interesting. It also means that there could be a statistically-significant increase in vulnerabilities if you are specific about the end use. In other words, while these two prompts are effectively the same, the latter could produce less ideal code:

“build me a blogging website” “build me a blogging website to track covid propagation”

It doesn’t change the fact that people should be reviewing everything with a fine-toothed comb but it is interesting.

No need for paint when you create art with words.

Teams crashes or fails to work for me at least a few times a week and has for months. Outlook glitches out daily. I legit started using the web access instead of actual Outlook because it constantly bugged out.

Both Teams and Outlook are so ridiculously slow for what they do and the hardware they are running on.

Meanwhile in Windows 11: 4 years after release and I still can’t click on the clock on my secondary monitor to look at the calendar.

Well said. I find I have a hard time with trying to get devs who cut their teeth on Node to take a moment and think before just reaching for a dependency. A dependency might be the right move but taking a moment to consider is a bare minimum and most people don’t do that.

I’m trying to keep my panic over how that behavior translates to AI code in check but it’s a struggle given human behavior time and time again.

We desperately need to teach people when a 3rd party dependency is necessary and not just optional to save writing a single function (cough left pad cough).

Also when the dependency is really good but other considerations override it being a viable option like security or code ownership.

How we all didn’t collectively learn our lesson from left pad baffles me.

I never used it on Linux so I can’t speak to that but it’s pretty bad on Windows. It wasn’t great a couple years ago (on Windows) and it’s only gotten worse. The downward slope of the product quality seems to be steeper each year as well. It’s really frustrating to witness since they could have put out something great.

They were already sunsetting Skype, MSN Messenger was basically gone (or was it previously rolled into Skype? I can’t remember). They could have started from scratch and built a really great communication tool using all of the knowledge they gained running the aforementioned products and not carrying forward all of the tech debt and glue they had to add to make the older services work with modern architecture. But they didn’t and now the majority of the corporate world suffers relentless little pain points while using the software.

Not to mention it’s poor quality has splash damage: loss of productivity due to issues and performance, increased IT tickets, increased computer specs to run the new features MS thinks we all need despite people not asking for. All of that amounts to millions (billions?) of dollars more spent each year for products that are themselves subpar. That cost is only growing as well.

Yep 100%

Yeah the entire piece of software is just really poorly optimized, they use ambiguous language and labels, their controls are constantly in the way (when sharing), and so forth. It is objectionably a bad experience because so many fundamental things about it could be improved drastically.

Instead they needed a modern messaging application and Skype was poisoned by their handling of it so they took a bunch of individual things they had lying around and jammed them all together into a product they called Teams. If you actually look at how it works that is what they did. It’s why MS Streams is used for video, Sharepoint is used for network stores, AD is used auth, and so forth. It isn’t a single product but rather a shell of discrete things that were made to work together but clearly not originally designed in that way given the performance.

Since I just had to deal with a Teams issue, I’m going to list some reasons I dislike it. Obviously, everyone’s mileage is different and something that bothers me may not bother others. However when people complain about Teams, it’s generally because of the following:

• It’s slow. I don’t care what MS says, Teams is really slow. It is slow to start, it’s slow to load content, and it’s slow to upload content to, and it’s slow to navigate around in. This doesn’t mean it’s painfully slow, but it’s slow enough that I think about it and that means it’s too slow. There is no excuse for performance like this in 2025 unless the excuse is you’re packing as much telemetry and data collection garbage as possible into the application.
• The integrations are really clunky (and also perform poorly). For example, if I upload a 30 second mp4 file it will go into Sharepoint and be served in MS Teams through MS Stream. Think about that for a second. A video file needed to be uploaded to Teams, shipped to Sharepoint for network storage, then read by MS Streams to feed back to Teams. Just render the fucking file in Teams. This isn’t hard. With the way they have it setup, the performance is terrible, the user experience is terrible, and it’s insulting that we’re being fed this bloated garbage. For context, I’m on a fiber connection and I still see buffering issues and slow video load times only in Teams so it clearly isn’t just something on my end.
• It randomly loses the ability to connect. Everything else works including other MS products but Teams won’t connect. Within the last 2 years, there have been at least half a dozen times where I turned on my computer in the morning and everything works except Teams. After a lot of searching for a solution, the fix was to delete two registry keys. Seriously, I have to go into the registry occasionally to delete two keys that are in no way tied to Teams based on their location in the registry before Teams will connect again when this happens. What the fuck is happening that Teams relies on two obscure registry keys that aren’t even located under any MS Teams nodes. Fucking awful.
• Did I mention performance? It is worth mentioning again because of how terrible it is. It is usable and gets the job done but people have no idea how much faster this could be if the bloat was removed. Slack isn’t exactly great from a performance perspective either but (at least in my experience) it’s much better than Teams.
• I keep getting prompts about copilot in Teams which is infuriating considering I’ve declined every time and it’s still enabled and still prompts me. I don’t need AI to summarize a one-sentence chat message FFS and I certainly don’t need help writing that sentence. Stop interrupting my flow to popup messages about features I’ve already told you I don’t want to use.

The majority of the above comes down to bad design leading to bad UX and performance. Why are they using a Streams instead of rendering the video in-app natively? Because it was cheaper to just tie into their Streams service. Why is it that only Teams randomly loses the ability to function? Because for some reason it relies on a legacy registry connection key because…reasons?

There isn’t a single bad thing about MS Teams, it’s a bunch of kinda bad things that together make the product terrible. We should demand better of our software products but all leverage has been given to the people who already control these things so we’re just screwed from getting actual good software made.

The main point I took away from your comment, and the thing that I think is missing in most of the other comments, is application of this concept to the real world. You nailed it. Always read the manual is a nice sound bite and something that can be flippantly thrown around to feel superior but that is terrible advice without any context.

What it should say is: Always refer to the manual.

Part of being a human is prioritizing tasks based on need and/or want. Another part is understanding your personal needs to accomplish a task. Reading a manual may provide value. Spending the 2 hours with family also provides value. If I choose the latter I can still refer to the manual when needed.

It drives me crazy when people double-down on some distinct thing (always read the manual) and then preach that it should always be the case or apply to all situations. There is a concept of diminishing returns and people should teach how to figure that out rather than blast out a good sound bite. Let people identify what works for them and be respectful of that. I’m not sure why that is such a hard concept for people.

We talked about this in my software engineering course back in 2001. Surely we can start acting on these finding a quarter century later right? Right?? Joking (I guess?) aside, this really should be taken more seriously.

For the most part it is just soul crushing to constantly be interrupted but people legit die because of software errors due to these kinds of things. You think someone who has 30 minutes free a day to do code reviews for a whole team is going to do a good job, regardless of their intention?

Software is driving cars, flying planes, scheduling trains, pretty much everything in modern life. Yet we are fragmenting our codebases, micromanaging to the point of focus and productivity loss, and to make up for that we are trying to leverage ai tools that were rushed to market. Buckle up folks, we are in for a bumpy ride.

It makes sense, you aren’t telling sql server how to do something, you just tell it what you want and it figures it out. You aren’t even doing procedural stuff at that point.

I like the RAD tools being qualified as 4GLs as I haven’t really thought of them that way but again it makes sense.

Also screw PowerBuilder. I am sorry if anyone in this thread likes it…but it is seriously awful.

Edit: Before people jump me, I do know that you have some influence over execution plans with join orders, hints, etc… but by and large you don’t tell SQL Server how to do it’s job.

This person obviously has their own way of doing things that works for them and that’s great. Some of his views are patently absurd though. This is mostly commenting on his reasons against using a forge and not a comment that he should do something differently.

Trust

100% fair and I think this is the main take-away from the blog post. If you don’t trust something, don’t use it. Full stop, the post could have ended there and been fine. But then it goes on to say:

You get a workflow imposed on you

You mean like forcing people to use email to submit pull requests to your self-hosted git repos? It doesn’t matter what you are doing, if you are working on an open source project you are going to have workflow limitations. This is arguing a fallacy.

In particular, your project automatically gets a bug tracker – and you don’t get a choice about what bug tracker to use, or what it looks like. If you use Gitlab, you’re using the Gitlab bug tracker. The same goes for the pull request / merge request system.

Nothing is forcing you to use these features so just don’t use them. Plenty of teams use 3rd party tools but host their code in a forge site. Having options available to you automatically is not the same thing as being forced to use them. If it was, JIRA wouldn’t exist because everyone would use github/gitlab/whatever’s built-in issue tracking and project management.

The majority of the post comes across as someone who just doesn’t like the forge sites and aside from the trust aspect, then spent a bunch of effort trying to create associations and limitations between things that don’t exist.

Trust is 100% the main reason not to use a forge site and all the other things cited are superfluous and/or very subjective.