# "Evil" 😉 antifeatures in the new forum?

**URL:** <https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288>\
**Category:** CPCWiki related\
**Tags:** features, cpcwiki, forum\
**Created:** [September 5, 2026, 6:03pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288 "2026-09-05T18:03:58Z")\
**Posts on this page:** 17\
**Page:** 1

<div class="post-metadata">

**Author:** ![cpcitor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/cpcitor/32/321_2.png) [@cpcitor](https://forum.cpcwiki.eu/u/cpcitor)\
**Post date:** [September 5, 2026, 6:03pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/1 "2026-09-05T18:03:58Z")

</div>

One of the things that I like with old machines is simplicity and parsimony I liked SMF because it felt lightweight and fast even on an old PC (or a modern machine at maximum powersaving settings).

When visiting the new forum I have noticed increased CPU usage and machine heat. Trying to assess whether it’s coincidence I’ve checked a few things.

1. uMatrix reports that the new forum wants to load content from [stripe.com](http://stripe.com) . Are you planning to include any kind of paid content? I guess not. In this case, perhaps try and find whether some useless modules are enabled? It might even have at least a small impact on server load and operation cost.

2. Both system and Firefox (about:processes) report some continued CPU consumption related to tabs showing the new forum. Firing up the developer tools show a timer and text that mentions **tracking the time users spend browsing**.

Is it a thing you voluntarily enabled? I don’t like the idea of a forum for retro enthusiasts that consumes CPU all the time and calls [stripe.com](http://stripe.com). Huge parts of the Internet are becoming enshitified and I understand that bringing new tools brings new costs. I’d appreciate if this was disabled.

Thanks!

 ![image](https://forum.cpcwiki.eu/uploads/default/original/1X/fd33d4bcba9c7f089f75bf788e6c7ff4b0c4f8cd.png)

---

<div class="post-metadata">

**Author:** ![Gryzor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/gryzor/32/7_2.png) [@Gryzor](https://forum.cpcwiki.eu/u/Gryzor)\
**Post date:** [September 5, 2026, 6:50pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/2 "2026-09-05T18:50:00Z")

</div>

Dude 😁

Stripe runs, unnecessarily and not transmitting anything really (because I don’t have a Stripe account above all 😂) because of the subscriptions feature. Subscriptions on Discourse are tied to Stripe, unfortunately. And, although I plan to use the Subs feature, it won’t be tied to Stripe. Unfortunately Stripe can’t be disabled without disabling Subscriptions altogether, but I have yet to think about how to implement it. It is, after all, still a WIP.

As for the tracking feature… Really? It bothers you that the forum tracks how much time you spend on it, like pretty much every other site out there? It’s not like it is used for ads or data mining and selling or anything.

Edit: oh, also, timers/trackers are used to decide whether to fire email notifications or live notifications in some cases. Also, it calculates whether someone actually read a post or just skimmed past it to award trust levels, remember their scroll position, and calculate topic reading times.

(Btw I’ve used the feature to weed out bots)

Nothing to be done about increased resource usage - that’s to be anticipated when you move from an antiquated platform to a new one. But I have to say it runs really smoothly on my secondary, ancient laptop.

That said, fair points, and thanks for the feedback. But calling them evil antifeatures is a bit too much​😁

PS I would really love to hear more of your thoughts.

---

<div class="post-metadata">

**Author:** ![Gryzor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/gryzor/32/7_2.png) [@Gryzor](https://forum.cpcwiki.eu/u/Gryzor)\
**Post date:** [September 5, 2026, 7:14pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/3 "2026-09-05T19:14:06Z")

</div>

Stripe/subs disabled.

---

<div class="post-metadata">

**Author:** ![Devlin](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/devlin/32/19_2.png) [@Devlin](https://forum.cpcwiki.eu/u/Devlin)\
**Post date:** [September 5, 2026, 7:18pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/4 "2026-09-05T19:18:23Z")

</div>

I understand the need for a bit of paranoia in this day and age of companies being _shady as fuck_, but I don’t think there’s any major concern.

In the case of here specifically, SMF has similar tracking features, so if it’s not been a problem for however many years of the old forum, why is it suddenly now, that we’re using a new forum software?

While SMF was great perf-wise, it was not resilient. It went down a lot.

The new forum seems to have a few features that are a bit pointless in my eyes. (why does a forum need achievements?) but i’d rather that, than no forum at all.

---

<div class="post-metadata">

**Author:** ![Gryzor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/gryzor/32/7_2.png) [@Gryzor](https://forum.cpcwiki.eu/u/Gryzor)\
**Post date:** [September 5, 2026, 7:26pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/5 "2026-09-05T19:26:38Z")

</div>

Achievements and gamefication are something I find a bit silly, too, but they’re tied to things like Trust levels so I’m not sure how to go about them. Personally, I ignore them.

As for performance, SMF was anything but. So you prefer a system that takes two seconds (or more) to render a page that then renders immediately (not always, because calls continued after that), or a platform that does work the client harder (it’s an Ember.js single-page app) but gives you everything without a hiccup?

(And let’s not talk about the maintenance that its “simplicity” required. You do have to take downtime into account. For instance, the reason we were able to get rid of cloudflare for most users was because of Discourse’s built in anti-crawler functions. Do you like your forum going down because Alibaba decided to knock it out, or take a bit of CPU and be up and running?)

---

<div class="post-metadata">

**Author:** ![cpcitor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/cpcitor/32/321_2.png) [@cpcitor](https://forum.cpcwiki.eu/u/cpcitor)\
**Post date:** [September 5, 2026, 8:00pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/6 "2026-09-05T20:00:11Z")

</div>

First, thanks a lot for your detailed answer. The details show that you have actual knowledge and command of the topics. Hence, all the more respect to you.

> Dude 😁

The ambiguity of written language… We’ve all remained chill all along, don’t worry. 😌

> Stripe runs, …

Seen you found in between a way to remove any call to the stripe domain, great!

> [@Gryzor](#):
>
> As for the tracking feature… Really? It bothers you that the forum tracks how much time you spend on it, like pretty much every other site out there? It’s not like it is used for ads or data mining and selling or anything.

As said in the first post, I was bothered about increased CPU consumption.

Jeff Atwood created Discourse. Just like for the StackExchange (StackOverflow etc) network that he also created, he took care that they are fully browsable without Javascript at all.

Yet Discourse looks “less minimal” and consume more resources. And unlike StackExchange and SMF has a completely different look when Javascript is fully disabled.

> [@Gryzor](#):
>
> Edit: oh, also, timers/trackers are used to decide whether to fire email notifications or live notifications in some cases. Also, it calculates whether someone actually read a post or just skimmed past it to award trust levels, remember their scroll position, and calculate topic reading times.
> 
> (Btw I’ve used the feature to weed out bots)

Ah those damn pesky bots.  
Thanks a lot for explaining.  
You have my trust and I guess the trust of the community.

> [@Gryzor](#):
>
> Nothing to be done about increased resource usage - that’s to be anticipated when you move from an antiquated platform to a new one. But I have to say it runs really smoothly on my secondary, ancient laptop.

It’s been like this for decades. Platforms grow and become slower. At some point, some people judge the slowness excessive and write from scratch a faster something, that starts the text iteration. 😁 🤷

> [@Gryzor](#):
>
> That said, fair points, and thanks for the feedback. But calling them evil antifeatures is a bit too much​ 😁
> 
> PS I would really love to hear more of your thoughts.

Answered above per your invitation, and below.

I still have a mental model close to the original web, that primarily serves documents. Programs (Javascript) are fully optional in the sense that whatever can work without code actually works without a Javascript runtime in the client.

Alas, most web developers don’t care at all and tend to just assume that CPU and bandwidth are unlimited, the visitors’s browser and the developer’s browser are the same, and all the web is 100% available and calls to third-party domain are free and natural. Why? Because it’s less effort than thinking about variety on the user side. The hard truth is the more you depend on, the more gets broken when something in another continent is down. And since breakdowns are rare, those fragile and bloated solutions get accepted and become the norm.

I like the CPC world where a byte or a CPU cycle is something you care about, not a negligible unit by several orders of magnitude.

Anyway, in order of importance:

- 🎉 the CPC community has its primary (in my opinion) website back online! I’ll be able to read from others, ask questions, mention my current work in progress, etc
- 🏆 you deserve recognition
- 🥳 this is great news
- 😌 Discourse allows to write posts in Markdown instead of fundamentally flawed Wysiwyg
- I suffer limitations like number of links or quotes in one reply, forcing me to cut off parts of this answer.
- … less important things
- … less important things
- … not important things
- 🧯 looks like Discourse consumes more CPU
- … not at all important things

---

<div class="post-metadata">

**Author:** ![Gryzor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/gryzor/32/7_2.png) [@Gryzor](https://forum.cpcwiki.eu/u/Gryzor)\
**Post date:** [September 6, 2026, 6:22am UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/7 "2026-09-06T06:22:47Z")

</div>

Long reply incoming for your -and everyone’s morning coffee.

To begin with: I want you -and every user here- to be happy with the result and the community. Is everything perfect? Heck no. But we’ve spent an inordinate amount of time over the summer planning and implementing the comeback of CPCWiki. I mean, think about having drinks next to a swimming pool and, instead of flirting with the girls at the next table, drawing up the server setup and backup nodes (true story). I actually spent most of my birthday implementing cpcjs here 😃

> [@cpcitor](#):
>
> First, thanks a lot for your detailed answer. The details show that you have actual knowledge and command of the topics. Hence, all the more respect to you.

Well. As I’ve said before, I’m no IT professional or anything. I work in marketing. When I decided to found the CPCWiki (first as a wiki, the forum followed) I knew little about it all. It is this very project that, over the years, that has taught me a LOT of what I know today and, unironically, has helped me with my professional work as well. I don’t think I know a lot, but at least I’m somewhere at the beginning of the Slope of Enlightenment on the Dunning-Krueger curve so that makes me cautious and forces me to learn things before acting.

I have a really soft spot in my heart for SMF. After all, I spent twenty years coddling it. But it _is_ an archaic solution.

Case in point: SMF is, in essence: “here’s a bunch of PHP files, they do stuff together, feel free to fuck it up”. And indeed, if you do screw something up, and have no backup to go back to, the official solution is to throw the installation package over the old files and revert the changes, losing pretty much everything you worked on. And that’s the upgrade route, too, which was the reason why with every big update we would lose all customizations and I’d have to work for days over something that should take minutes.

I guess they’re taking care of it with version 3 which may come by 2030… or not.

Discourse, on the other hand, says: “here, fill in this app.yml, then sit back or go for a looong walk while we download and configure and install a whole compose stack. You can have a look, but believe us, you don’t want to”.

So SMF is pretty much more open, and works fine out of the box, but breaks very easily (and some of the design decisions are questionable). Discourse is very opaque in that regard, but it does expose a few entry points that allow you to do stuff without breaking it.

> [@cpcitor](#):
>
> The ambiguity of written language… We’ve all remained chill all along, don’t worry. 😌

😉

> [@cpcitor](#):
>
> Seen you found in between a way to remove any call to the stripe domain, great!

No, there was no way, I disables subs altogether. At least until I figure out how to implement it.

> [@cpcitor](#):
>
> Jeff Atwood created Discourse. Just like for the StackExchange (StackOverflow etc) network that he also created, he took care that they are fully browsable without Javascript at all.
> 
> Yet Discourse looks “less minimal” and consume more resources. And unlike StackExchange and SMF has a completely different look when Javascript is fully disabled.

Well. But Atwood (and the other guy, whatshisname) pretty much felt that static pages were a thing of the 90s, and that’s why he went the other way with Discourse. Sure, a SMF installation will render just as fine on a PC from 2005, but this doesn’t say much when you take functionality into account. For instance, I’ve started this comment on my laptop and I’m continuing on my phone - and I’ll go to my laptop again later. Oh, and I seem to remember people complaining about the [ASP.NET](http://ASP.NET) MVC back in the day as well 😃

> [@cpcitor](#):
>
> Ah those damn pesky bots.  
> Thanks a lot for explaining.  
> You have my trust and I guess the trust of the community.

Much appreciated. Let me reiterate, the data never leaves the server (I mean, sure, Discourse pings here and there and GA runs in the background, but it’s not like I do anything else with said data). I actually feel underappreciated that nobody has approached me to buy the dataset 😃

> [@cpcitor](#):
>
> It’s been like this for decades. Platforms grow and become slower. At some point, some people judge the slowness excessive and write from scratch a faster something, that starts the text iteration. 😁 🤷

Well said. But, I really feel Discourse is light enough. I have two tabs open right now, and with my laptop (again) CPU pinned down at energy saving mode Vivaldi tells me it takes precisely 0% CPU and about 350MB. That’s good enough, I reckon. Ok, the CPU is a Core Ultra 7 255HX so even when running at 2.4GHz it’s powerful enough, but…

> [@cpcitor](#):
>
> Answered above per your invitation, and below.
> 
> I still have a mental model close to the original web, that primarily serves documents. Programs (Javascript) are fully optional in the sense that whatever can work without code actually works without a Javascript runtime in the client.

Yeah, the key is “whatever _can_ work”.

> [@cpcitor](#):
>
> I like the CPC world where a byte or a CPU cycle is something you care about, not a negligible unit by several orders of magnitude.

Not that I disagree with you, honestly, but we could run this as an ASCII BBS and see how much people would use it 😉

> [@cpcitor](#):
>
> - I suffer limitations like number of links or quotes in one reply, forcing me to cut off parts of this answer.

This has to do with trust levels. I upped the newuser limits a bit, but I think they were ok before, too except for rare cases:

 ![image](https://forum.cpcwiki.eu/uploads/default/original/1X/60c51d537ecf1060d5a837a76f08994d23e050a9.png)

I’ve upped your Trust Level manually.

Also, the quotes limit is a global variable - a design decision to avoid a chaos of nested replies. I’ve upped it from 5 to 10 (because I was also hit with it here 😃 )

Anyhow, really interesting discussion. Thanks for your kind words 🙂

---

<div class="post-metadata">

**Author:** ![Devlin](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/devlin/32/19_2.png) [@Devlin](https://forum.cpcwiki.eu/u/Devlin)\
**Post date:** [September 6, 2026, 8:21am UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/8 "2026-09-06T08:21:26Z")

</div>

> [@Gryzor](#):
>
> Not that I disagree with you, honestly, but we could run this as an ASCII BBS and see how much people would use it 😉

I’m in, so long as it’s CPC compatible!

---

<div class="post-metadata">

**Author:** ![cpcitor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/cpcitor/32/321_2.png) [@cpcitor](https://forum.cpcwiki.eu/u/cpcitor)\
**Post date:** [September 12, 2026, 10:48am UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/9 "2026-09-12T10:48:21Z")

</div>

> I liked SMF because it felt lightweight and fast even on an old PC (or a modern machine at maximum powersaving settings).

After some experience on the real new thing 🥁(drumroll) … now I like Discourse!  
Excellent experience on mobile. Fast and all.  
Hurray to @Gryzor !

---

<div class="post-metadata">

**Author:** ![mattl](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/mattl/32/174_2.png) [@mattl](https://forum.cpcwiki.eu/u/mattl)\
**Post date:** [September 12, 2026, 12:43pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/10 "2026-09-12T12:43:05Z")

</div>

I’ll buy the dataset for $1 😉

---

<div class="post-metadata">

**Author:** ![Gryzor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/gryzor/32/7_2.png) [@Gryzor](https://forum.cpcwiki.eu/u/Gryzor)\
**Post date:** [September 12, 2026, 1:55pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/11 "2026-09-12T13:55:33Z")

</div>

> [@cpcitor](#):
>
> After some experience on the real new thing 🥁(drumroll) … now I like Discourse!  
> Excellent experience on mobile. Fast and all.

I’ve had my eye on Discourse for quite some time now, but of course migration was/is difficult to impossible. Glad to hear people are liking it!

---

<div class="post-metadata">

**Author:** ![cpcitor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/cpcitor/32/321_2.png) [@cpcitor](https://forum.cpcwiki.eu/u/cpcitor)\
**Post date:** [September 12, 2026, 2:08pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/12 "2026-09-12T14:08:13Z")

</div>

It was impossible. @Gryzor did it. 😎

---

<div class="post-metadata">

**Author:** ![Gryzor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/gryzor/32/7_2.png) [@Gryzor](https://forum.cpcwiki.eu/u/Gryzor)\
**Post date:** [September 12, 2026, 4:48pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/13 "2026-09-12T16:48:26Z")

</div>

Yeah, nah, it was a lot of work but not nearly impossible, honestly.

---

<div class="post-metadata">

**Author:** ![SerErris](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/sererris/32/227_2.png) [@SerErris](https://forum.cpcwiki.eu/u/SerErris)\
**Post date:** [September 15, 2026, 4:33pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/14 "2026-09-15T16:33:06Z")

</div>

With APIs now available it could be possible to import threads from the old forum with an AI agent. But to be honest I am not sure of the value. However you probably do not want to continue running two boards and I would love the old material not going away. All the knowledge in the old forums. That is how I learned a lot on the Amstrad (relearned) and things I never knew in the first place.

So my primary goal on that would be preserving all the knowledge.

---

<div class="post-metadata">

**Author:** ![Gryzor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/gryzor/32/7_2.png) [@Gryzor](https://forum.cpcwiki.eu/u/Gryzor)\
**Post date:** [September 15, 2026, 7:30pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/15 "2026-09-15T19:30:23Z")

</div>

> [@SerErris](#):
>
> With APIs now available it could be possible to import threads from the old forum with an AI agent

No it wouldn’t, actually, be possible; what about the users? PMs? Lots and lots of details that aren’t possible through the API because that works on the surface…

---

<div class="post-metadata">

**Author:** ![SerErris](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/sererris/32/227_2.png) [@SerErris](https://forum.cpcwiki.eu/u/SerErris)\
**Post date:** [September 15, 2026, 7:53pm UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/16 "2026-09-15T19:53:22Z")

</div>

Hm… I see the issue and I do not have a good solution right now. I will think on it a little bit.

The idea would be to let an AI agent just read the old forum (which it can right now - no API needed) and from there post it to the new forum. Maybe only curated things …

I just searching for ways to preserve that knowledge. If the old forum goes away, dies, etc. this information is lost forever.

But again - I do not have a good idea on how to preserve it in a good way in a modern archive. Let me think and chew on it a little bit.

---

<div class="post-metadata">

**Author:** ![Gryzor](https://forum.cpcwiki.eu/user_avatar/forum.cpcwiki.eu/gryzor/32/7_2.png) [@Gryzor](https://forum.cpcwiki.eu/u/Gryzor)\
**Post date:** [September 16, 2026, 7:57am UTC](https://forum.cpcwiki.eu/t/evil-antifeatures-in-the-new-forum/288/17 "2026-09-16T07:57:57Z")

</div>

Yes, that’s what I meant with “works on the surface”. It would be trivial to crawl the old forum according to some criteria and create new posts here through an agent. However, the first thing you’d miss (and it quickly unravels from there) is that EVERY post and thread would be under the same user name, causing a chaos.
