OSDev.org

Hi,

Does anyone else miss the old quality of this forum?

Especially when it was at mega-tokyo forums
and also the first times at osdev.org?

What happened to that community?

Regards.

What's wrong with now?

@m3: Stop complaining and write some quality stuff than. Without even one contribution here your opinion means nothing.

I'm sure one of the staff can check the IP and link it to several of the other dozen m## accounts out there.

But yes, times change. People nowadays write something java or C#, think it was easy and then claim they mastered programming. The skills that were abundant in the older years are no longer taught to the current generation and that is becoming rather obvious in the ever increasing gap between levels. The actual questions that are being asked are rather boring and hence are the answers.

Fact is that the current standard means you can have a CS degree and still fail to have even one of the points mentioned by the required knowledge rule covered.

What kind of topics would you rather like to see?

Those that don't involve people trying things far above their actual capabilities.


Seriously, actual development is a dead subject. If you can actually program you don't need to ask those questions - it's all way too much documented. What's left is actually doing design or reverse engineering, or stupid questions.

Yo:

Regardless of the direction that the forum moves in, the admin of the site has stated that he wishes to see the focus of the website as a whole move from the forum to the Wiki.

There is a fair amount of "quality" in the wiki. I encourage you to go forth and increase the quality ever further.

Additionally, the maturity of the community members has increased, along with the base of core knowledge since mega-tokyo. I was not a member of this forum during the mega-tokyo days, but I have read a large number of posts from that "era" as part of my research. At one time I read, beginning at his first post ever, and proceeding onward, a large number of old posts from Brendan, so I have some idea of the "level" the community was at.

Things that used to mystify at that time are now "beginner mistakes". Things that mystify now will soon also become "beginner material". For example, the once very arcane subject of Non-uniform memory access is now a mostly unimpressive subject, with several members in the community having claimed to have added support for it in their designs. In the mega-tokyo days, the posts on topics like the LAPIC/IO-APIC chips were hot fire, and seemed to be the real meat for the big minds.

Now the APICs are old news, and topics like the first implementer of a UDI environment for the further progression of the community are what incite the most discussion. I don't think that the "quality" of the discussions has deadpooled, but rather, the topics which used to be really advanced at that time are simply too trivial now, and too well understood to merit real discussion, and the questions that are raised by the majority of the general membership don't really incite the same level of excitement that they used to, so it seems as if discussion has died down.

In reality, the knowledge has increased so much in the "core" membership that most questions can just be answered rather than discussed. Is this a good thing? Obviously, yes.

--Peace out
gravaera

Why do I see this same tired old question on every forum I visit? It's so predictable.

The good old days were no better than the good now days.

I see this a bit differently and don't think it's a bad trend. Yes, the percentage of good programmers has decreased. But that's only because programming is now available to many more people, of which not all need to be expert programmers to reach their particular goals. I'm glad that more people can take advantage of the flexibility of computers, even if they may not know how to define computation, for instance.

I agree that there is nothing wrong with the current quality. I also did some research and I read through old discussions. However, now I am starting to realize how advanced the knowledge level is at best. I have to be careful with my posts. If I were very excited of having written just a working operating system (rather simple, of course), it would not be even worth mentioning here. It is so an already done thing.

I could not help but notice that maybe the visible OS developing progress slows down the further you get. It may take a year of coding but your OS looks and even behaves exactly like the original one. The small "invisible" improvements and the general elegancy of code become more and more interesting.

I feel that I am right on schedule when it comes to learning but the goal is unknown. Perhaps the goal is not so important because it goes farther all the time.

It's a Long Way to the Top (If You Wanna Rock 'n' Roll)

Somebody who cut their teeth in the days of MS-DOS would have learnt assembly language, possibly even before C, and done a fair amount of systems programming, without even realising that was what they were doing. They would have known what an 8259A was, and that gave them a head start when it came to osdev.

Nowadays people want to know if an operating system can be written in PHP, and, if not, why not.

Probably the point of it all is that new users who know something (and the frequent comments from experienced ones), but mainly more experienced new users will have to help other new users who know much less.

As in, the responsibility to do the teaching job of past years will have to be done spontaneously by good new users; the already-established users would more likely get bored at having to repeat their past teaching job (I also get bored at those situations, anyone would, after having other things to do at other higher level).

New users must also learn and be given help to learn asking questions beyond their immediate problem; in many cases it is difficult to help them in such a tiny, isolated scale; and they would learn much more if helped constantly over time.

The "original users" as well as people who already know this subject and their individual pieces will find every question boring given that, after all, this forum is intended for people who is just starting in the subject and/or that is trying to learn more about it in practice.

But what must be kept in mind is that the people from the new generations will always start out ignorant. They have the disadvantage that this kind of knowledge is used only in the most specialized segment of computers (places like the system departments from Microsoft, the Linux kernel maintainers, etc.) and thus they are never provided formally with this knowledge.

Even if they have the potential of the best programmers, they will need help to master an environment that is different from what surrounds them.

And there are users from countries where OS development is just non-existent. It is bound to happen thanks to Internet globalization.

For them an their own peers, there is no single piece of (rich) information that isn't amazing, and will need much more help to learn it given their considerably more limited knowledge; it will always keep happening as long as new generations of programmers keep appearing.

Probably every time a question like the one of this thread pops up, it signals the arrival of one or more persons from a newer generation. Rightly and genuinely new for them, but something that older ones have seen several times in the past already.

__________________________________________________________
__________________________________________________________
__________________________________________________________
__________________________________________________________
But we are intelligent enough to bring the quality back. It's not just about low-level details but also about including things that make any "hobby" OS useful, like image and document viewers (could be deveoped in the General Programming Forum).

But it probably requires for a few fellows to be less elitist. Am I saying that MegaTokyo was less "elitist" and more cooperative than the current situation? Of course, absolutely, even as recently as 4 o 5 years back, this forum was much more cooperative. The important thing is that now, any of us who has something good to share, about code and documentation of any technic kind, can revert it very easily.

Things are far from being already lost.

Personally, if you ask me about low level programming I can give some basic help, but if I can work together with others, with feedback, then I can develop good code very rapidly, and it doesn't become tedious, because I know that what I'm doing is immediately useful (when instant gratification is otherwise a non-existent concept in truly lone OS development, or any other complex development area).
__________________________________________
__________________________________________
__________________________________________
__________________________________________

I miss the quality of every olden media, like the MegaTokyo forums. There was much more cooperation and rich information back then. Now even paid material is of a much lower quality (no true technological/algorithmic roots are taught, much less a proper tutorial, only encyclopedic packs of information; and as we know, there is a long way between how the raw specification looks like and how the actual tutorialized explanation and practical implementation looks like).


I learned Assembly language well after MS-DOS was the OS of choice (XP age).

But I always wanted to learn it, and I use it even to turn files in a given file format to a human-readable ASCII format.

The problem is not the language you choose to use, provided you know what you are doing. And also the general lack of cooperation Internet-wide (it is better to buy books to learn than to expect to get it all from the web; then you can share tutorials and code, and add to what lacks the most in the wild).

For instance, my personal choice is to program the base of my kernel in Assembly (and C, and also a modified version of C/Assembly I created).

But after that, the first scripting language I would seek to implement would be JavaScript with LLJS (Low Level JavaScript) capability (typed arrays, etc.), and also HTML5.

With that, I would have a framework to create GUIs in a standard, industry-proven way, and reuse a lot of code. And it would be cross-platform portable.

Of course, the interesting thing is to implement it. And of course, there is a long way from a basic kernel to a "web-enabled" operating system, with a lot of file format readers (GIF, PDF, PNG, ZIP, ISO...).

______________________________________
______________________________________
______________________________________
______________________________________
And yes, my programming choice is to always use Assembly language, but also use the so-called Low Level JavaScript along with HTML5 (and some server-side with PHP), to make it possible to use the capabilities gained in OS development, in the most popular and platform-dependent, standard environment, which is the modern web.

Both programming environments can work as a reciprocal feedback, and that accelerates learning really tough concepts at the OS development side.