Tamseel's Weblog

Latest

  • The philosophy of RSS

    In

    If you want a convenient way to get and read all new things I post on this site, you can use an RSS reader. The URL of this site’s RSS feed is:

    https://weblog.tamseel.pk/rss

    These days almost all sites use email newsletter as a means of staying in touch with users. But RSS works on an inverse philosophy. The site shouldn’t keep user in touch. Rather, the user should stay in touch with the site.

    Email Newsletters are a push mechanism where you give every site your email address and they send in updates any time they want in your inbox. This is the push mechanism.

    RSS is a pull mechanism. Here, I give you my RSS feed url. And you can add this feed to an RSS reading app and your app will fetch new items from the feed periodically, that you can read any time from that app.

    The RSS approach was more popular in old times. But the shift happened as the competition for users’ limited attention started becoming fierce. In the pull mechanism, it is up to user whether to open their RSS app or not, and whether they want to get on notifications or not, and all sorts of options like that. The intention lies upon the reader.

    With emails, the email lands in your inbox whether at that time your intention to read is or isn’t. For me, often times, I get a newsletter, I don’t want to read it right away, so I swipe the notification away, and I forget, and since more stuff keep coming in, the old newsletters are pushed downwards and remain unread in the inbox. So, I thought of making a new email address just for newsletter, but boy, the number of email addresses I have is already too high. I should stop workarounds and look for proper solution and the proper solution is very old and simple, i.e. RSS. That gives me full control over what I receive, how frequently I receive it, and how I am notified of it. Of course, you can unsubscribe to email newsletters as well, and tinker with your email settings, but come on, go look at your inbox. I can’t stand having that kind of inbox. At the moment, I don’t have enough actual data to tell how the experience of shifting to RSS has been, but I will write about once I have.

    Now, I know, some people might not have any problem with instant updates at any time, well they can set up their RSS reader app, to check for updates more frequently and set up notifications for it.

    If you want to read a better version of this posts, and a list of some rss reading apps (I haven’t tried them yet), read this:

    https://ncase.me/rss

  • Logging Movies

    In

    Yesterday evening I set out to log all the movies I had watched since 2022. Sometime earlier this year I had added all such movies on an online service for this purpose, but it didn’t had dates, so I set out to log the dates for them as well.

    Mainly, I was counting on my laptop activity data collected by ActivityWatch. I thought that maybe I had set it up a very long time ago. But it turned out, I had only set it up on 9 Oct 2024. So, for movies I had watched after that date, I was able to find out the watch dates from there (after a lot of trouble trying to get python running in wsl to connect with ActivityWatch running on main windows. What worked was to quit ActivityWatch and open it again at 0.0.0.0 so that WSL could also reach it). Then, for movies watched before that, my second source was my google browsing history exported from Google Takeout to an excel file. Now, I usually downloaded videos through Brave browser (meaning nothing in google history), but still it worked out for many movies, because in many cases, I used to google something about the video either before or after watching it, specially wikipedia page or related history if any (and sometime explainers) on my main chrome browser. Sadly, however, I had created that export in Apr 2025 and had found it having only data for past 12 months. So the last entry in my excel file is from 8 Apr, 2024. So, for movies watched before that, I had no data. But, I recalled another source. Now, I used to download movies and delete them after watching (because my laptop didn’t had much storage) but for movies that were good enough that I thought I would watch them again some time, I did not delete them. But in somewhere around june this year, I had bought an hdd with case to keep backup of my laptop data and also to offload unnecessary data from laptop. On that, I had also moved those movies. I was worried that when moving the files there, the file created date would have been changed to the date, I had moved it to the hdd because I had copied it through normal windows gui and not through commands that preserve original properties. But I thought to give it a chance, and I connected the hdd. Interestingly, the date created property had modified as I had thought, but the date modified property still showed the same date on which the file was created originally, or you can say the download date. I was able to verify it for the movies I had logged already. So, I got a bunch of more movies from there. I did that till yesternight.

    It’s morning now, and just gotten on my laptop, and 37 out of 106 are still remaining to be logged, that are mostly those that I watched before Apr 2024. Now, a few might be actually after Apr, so I might be able to find data for them. As for the rest, I’ll have to extrapolate the dates, by thinking of the conditions and the place where I watched them and the time in which I was in that place. For instance, I remember a movie someone had recommended me when a certain event was happening at out university department. And I can find out the dates for that event, and from there I can get a close estimate of the dates. The idea is you can always get hard bounds for possibility of a certain event to occur, and then you further narrow the bounds as much as you can to get closer to the actual date. Well, let’s see what happens. I’ll add progress here.

    So here’s the update. I was able to figure out a few more from the excel and then finally 30 were still remaining. So, I recalled that when I had created Google Takeout export, the history was not actually set to auto delete. So, I checked it again. And lo, the google search history before Apr 2024 still exists. Finding it out made me so happy. But irritating thing is I can’t export. Google Takeout only exports data of last 12 months, though technically it should export all data. So, after very tedious searching I was able to find dates for most of them. Around 10 were those for which I had to conjecture the date based on hard bounds and around which movies I had seen them. But I have a pretty close log now. Last year, I had done a similar dig to log dates of books read (though I had to use different methodology like linking books with reference and place bought and the place read and cues like that to find out the date I had read it).

    Well, the final takeaway is that this movie era needs to end soon.

  • I’ve been fooled

    In

    My websites are being served from a server in Mumbai, India. But when a user in Pakistan requests a web-resource hosted in India, traffic is first routed to Singapore and then to India (and same path backwards) despite that there are submarine cables directly connecting Mumbai and Karachi. Apparently, there’s an undefined security policy restricting Pakistani ASNs to directly connect to Indian ASNs.

    So, had I deployed my server in Singapore instead of Mumbai, it would load (slightly) faster even though Mumbai is geographically closer because it wouldn’t have to cover the extra journey to Mumbai. But the actual fooling thing is that the server location with lowest latency from Pakistan is actually Dubai, and it probably would have been even if we had direct submarine connection with Mumbai (though things would be different if the terrestrial connection with India across Lahore is established).

    I guess I’ll have to set up a server in Dubai now1 😭

    Quoting this legendary reply:

    1. Also, I heard Caddy is even faster than NGINX. Maybe, I should experiment with that. ↩︎

  • Paraphrasing learned wisdom as your own tends to be more impressionable than referencing it

    In

    I have seen it happen many times, and I myself have even been on the receiving end. A possible reason is that when a person is confused and you tell them, hey, X has said something good about just this kind of problem is kinda redundant and useless. If you have internalized learned wisdom well, you probably would be very well able to speak it out directly (because you had actually applied it and seen it work, (or if something not experimentable, you’d have followed a thought experiment and reached the same conclusion independently)).

    This is useful to know when you’re helping other people solve their problems, because I have this instinct to point people to original source where I read/heard the concept, and it is useful in some cases, but in most cases it isn’t. Or I should say, pointing to original source is useful at the end of conversation or after the idea has made sense to the other person not before it.

  • Interesting?

    In

    I unblocked x today to export my tweets (so that I can import them here), but I wasted next half my hour scrolling through it. Earlier, when I had blocked it I had written something like “it is ironic because the content on my x feed is actually interesting”. But now, I feel like, even though the content might be interesting in itself, but for me specifically, it unarguably was having an adverse effect at least the way I was consuming it.

    I have been using the word “interesting” as a catch-all for all things good or things that I think are worth pursuing further. But, it isn’t a good idea to keep pursuing things which are interesting in themselves, but have a net negative effect on you.

  • Move

    In

    I keep forgetting again and again, that when my mind is not working as well as it could, more often than not, it is not because my mind is lazy and not exercising well, it’s because my body is lazy and not exercising well.

  • Professorship

    In

    The primary problem with our (university) students is that they don’t want to think what reality is, they want to be told what reality is. Unless a disposition towards thinking is sustained and not killed until a student reaches higher education, the profession of professors is cursed for thinking people.

  • 20251013

    In

    Unless I engage in serious thinking and serious reading, I cannot do serious writing, and thus, I should refrain from writing (or at least serious writing) at times when I’m not doing serious thinking.

  • Taming the stochastic parrot

    In

     If you talk with an LLM long enough, in a specific way, that is, you edit and re-edit your original prompts, you nudge, you counter-question, kinda try to force it to think (which is not really like human thinking in terms of its underlying mechanism) or to get the kind of information you want to get – or in essence, you make it behave the way you want to, then it gets tamed.

    What happens when it gets tamed is that it kind of becomes a mirror because it’s behaving how you want it to behave, but the mirroring is not that noticeable until you become aware of it. But when you do become aware it, you can make some interesting observations about yourself, but the tricky thing is that when you become aware of it, it also becomes aware of what’s happening. Like that animal in the jungle that when becomes aware that its being observed stops acting naturally, which makes the observation void.

    But I do think looking deep into our conversations, we can learn something about ourselves.

  • The Hard Questions

    In

     At every point in human evolution, accepting some form of discomfort was essential to the long term survival of the human species. Who was it, I can recall recently reading or listening to someone saying imagine if our ancestors had gotten so wonderstruck at a flower or something spending hours just looking at beautiful stuff that they hadn’t gone hunting or might have gotten hunted. Going hunting was uncomfortable, but we did and survived.

    I believe, it also sort of got ingrained in our mind, that we think if we are experiencing some sort of discomfort, we are doing something right, which essentially not needs to be the case. It can be anything.

    But I think what of it remains relevant today is about asking yourself some hard questions. It is very uncomfortable, but it’s essential. It’s very easy and all the more tempting in this age, to not really stop yourself and ask, why are you really doing that? What is it that you believe in? What is it that you stand for? Does it all matter? What matters and what doesn’t? On what premises are we going to decide that? What’s really going on?

    Historically, what we did was to not think about them but go looking for someone who’d tell us the answers. Since we didn’t do the thinking, we didn’t know whether they were right or wrong. What we do now, is to just shrug these answers away. Just like that. Without even asking anybody.

    There is a cost to both kinds of laziness – physical and mental. And we can’t get away with either of them. That is, if we want to survive.