[sid] yeah, i broke it...

[pylkko]

..or roll back to yesterday's snapshot

[anticapitalista]

RTFT (Terninal) - a must for those of us using sid. sid can be ruthless, you mess up just once, and .. kaput!
As they say (whoever they are) 'Patience is a virtue ...' (I'll let you google the rest and the source)

[pylkko]

Well, I remember when you said in the backup thread that you use this method, and I thought that it sounds like really good idea. Primarily because, well, there is little effort involved compared to full system backups and systems don't brake that often if you know what you are doing. Also, reinstalling every now and then you can make other changes to the OS that might be hard or impossible without starting from scratch. However, I never thought much about what would happen when packages in the repo no longer correspond to the ones in your dumped list. Intersting. I think I would use this method for a machine with little packages and on stable (like a home server or something) but maybe not for a dev OS system.

I currently run a Sid system on btrfs and one clear advantage is that a snapshot can be made in less time than a dpkg dump. But of course, if you consider LVM "too new" then... I have been using it on a daily basis for something like 6 months without problems.

[dasein]

...what does make me a little nervous is Yet Another Layer between my data and userspace.

Couldn't agree more. (Huge surprise, huh?)

I've never been able to convince myself that LVM adds value commensurate with its complexity.

@Wiz10K: You definitely got the skillz, so if this is your first time breaking Sid, then you're simply not trying hard enough.

[pylkko]

two small things:

1) when I said that I hve run Sid on btrfs for 6 month without problems I meant it only with respect to btrfs...

2) I have never and probably will never use LVM because of the extra unnecessary complexity

[No_windows]

[quote="pylkko"
2) I have never and probably will never use LVM because of the extra unnecessary complexity[/quote]

What's complex about it? I've installed Debian with LVM, and have never encountered any kind of complexity, but I've never tried to change volume sizes or really used it for anything, either. Am I slowing my computer down by using it?

[dasein]

I've ... never encountered any kind of complexity, but I've never ... really used it for anything, either.

In other words, you're saying it's superfluous?

To paraphrase Strunk and White's Elements of Style (with deepest apologies)

Good technology is concise. A technology should contain no unnecessary parts, a program no unnecessary routines or dependencies, for the same reason that a drawing should have no unnecessary lines and a machine no unnecessary parts.

It's a philosophical thing. (Anyone truly interested in such deep philosophical questions should read Winograd & Flores' Understanding Computers and Cognition. It will change how you think about technology.)

(Sorry for the hijack.)

[Bulkley]

To paraphrase Strunk and White's Elements of Style (with deepest apologies)

Good technology is concise. A technology should contain no unnecessary parts, a program no unnecessary routines or dependencies, for the same reason that a drawing should have no unnecessary lines and a machine no unnecessary parts.

Agreed. Remember the Commodore 64? It had a word processing program called Paper Clip composed of some of the finest programming ever. It had to be; on a 64 there was no room for slop. As processors got faster and drives and memory got bigger programming got sloppier.

[No_windows]

I've ... never encountered any kind of complexity, but I've never ... really used it for anything, either.

In other words, you're saying it's superfluous?

For me, probably. I'm a fairly basic desktop user, However.

[millpond]

Personally, I never delete an installed package, at least not until I am happy with its successor (which I then archive in its place).

It also allows you to extract the contents directly, when APT gives hissy fits.

[sunrat]

siduction forums Upgrade Warnings is a great resource for Sid users. Often someone will have a breakage and post it there so you can hold off until an all-clear is given. Of course you may be the first to experience a d-u problem, then it's polite to post and warn others.
I visited there earlier and Sid has several transitions happening currently, so upgrading may be perilous.
https://forum.siduction.org/index.php?board=22.0

[edbarx]

To paraphrase Strunk and White's Elements of Style (with deepest apologies)

Good technology is concise. A technology should contain no unnecessary parts, a program no unnecessary routines or dependencies, for the same reason that a drawing should have no unnecessary lines and a machine no unnecessary parts.

An idealisation of what technology should be in the eyes of an idealist who hasn't yet realised how complex reality is. I expect a quote like this from an intellectually gifted young adolescent, but definitely, NOT from an adult past the age of thirty.

Living cells work, yet internally, they are extremely complex. Even a DNA has redundant genes which further contradict this idealist philosophy.

According to the quote, a DNA is a bad design. Living cells, are nanotechnological systems that occur naturally.

[millpond]

To paraphrase Strunk and White's Elements of Style (with deepest apologies)

Good technology is concise. A technology should contain no unnecessary parts, a program no unnecessary routines or dependencies, for the same reason that a drawing should have no unnecessary lines and a machine no unnecessary parts.

An idealisation of what technology should be in the eyes of an idealist who hasn't yet realised how complex reality is. I expect a quote like this from an intellectually gifted young adolescent, but definitely, NOT from an adult past the age of thirty.

Living cells work, yet internally, they are extremely complex. Even a DNA has redundant genes which further contradict this idealist philosophy.

According to the quote, a DNA is a bad design. Living cells, are nanotechnological systems that occur naturally.

Not quite true. Despite the extrme complexity of some DNA, often largely derived from viral and other sources - organisms do prefer simplicity, and unnecessary functions will after a period be lost. To minimize complexity.

Think cave species that lose eye and pigment.

Think reason why we cannot digest cellulose. Or make all our fats and amino acids. Like bacteria do.

Keep in mind also that introns, long believed to be useless because they do not encode genes, may actually be a means for modulating, or even frame shifting gene expression.

This is a key omission in the theory used to create GMOs, and why they may be ultimately quite dangerous.


So the complexity in the human genome may actually be useful, but Entropy can also apply to biology, species can and do degenrate over time (regression to infantile forms - snakes and humans).

[pylkko]

An idealisation of what technology should be in the eyes of an idealist who hasn't yet realised how complex reality is. I expect a quote like this from an intellectually gifted young adolescent, but definitely, NOT from an adult past the age of thirty.

Living cells work, yet internally, they are extremely complex. Even a DNA has redundant genes which further contradict this idealist philosophy.

According to the quote, a DNA is a bad design. Living cells, are nanotechnological systems that occur naturally.

I disagree, with your point, but also with your out right rudeness. It's really uncalled for.

Most people, save some real "whack jobs", would argue that DNA was not designed. Furthermore, non-coding genes -and that's what I believe you are referring to - are by no means "redundant". And if we want to be pedantic, it certainly is not the case that DNA "has genes".

The only real problems that I see dasein's norm is that I would maybe include the clause "to the extent that it is possible and feasible to do in the time and resource limitations present". Sometimes (often) refactoring/redesigning a design just to get rid of some unnecessary or now obsolete elements is just not worth the ordeal. Also there is the problem always related to norms: who get's to decide and on what basis if something is "necessary" or not. And maybe thirdly, sometimes designs might have so called "preadaptive" non-functional elements. That is, elements in place that currently have no use but which attempt to preempt some future use (like some API element in a library/program).

[edbarx]

Most people, save some real "whack jobs", would argue that DNA was not designed. Furthermore, non-coding genes -and that's what I believe you are referring to - are by no means "redundant". And if we want to be pedantic, it certainly is not the case that DNA "has genes".

You are right about rudeness, but you should remember this dasein, is not impeccable... rudeness is in their blood.

Of course DNA was not designed: it evolved, but that does not exclude it from using the laws of physics, mechanics and chemistry which happen to also be used in technology. The DNA, has non-coding genes, and in agreement with your view, research is indicating these are genes that may be used in turning on and off coding genes.

What I said about the DNA, is that there are protein coding genes that are redundant. In fact, as an organism, we humans, get two complete copies at conception. So, half of them are technically dormant unless they happen to be passed forward to the offspring from both parents.

This clearly contradicts the quote posted by dasein. And yes, it is idealist to assume everything must be self contained and simple. Such idealism shows a detachment from reality which works and is very complex.