This is a category error. A cell is not a thing that has a goal. To imagine it has one is pure anthropomorphism. The religious may have other views of course.
Just because things interact doesn't mean there was a goal end state.
But, life is different. Life that survives better, out competes life that doesn't survive as well. So biology becomes incredibly fine tuned for one goal: survival.
Not survival of the individual, although that is part of it. But survival of the thing that makes the choices of how to survive. I.e. our genes. They evolve to enhance their own survival.
Which is why we care about our family more than others. Genes for caring about family have an advantage over genes that don't. Because many of our genes will also be in our other family members.
So from that perspective, each of our cells cares about doing its part to help ensure our genes survive. So it cares about itself surviving, but it also cares enough (when working properly) to sacrifice itself when it is too damaged. Because the cell doesn't last forever no matter what, but the individual it is in can pass on its genes if it sacrifices for that individual.
> So biology becomes incredibly fine tuned for one goal: survival
Again, this is a category error. It works well as a metaphor, but evolution has no goal. Evolution is a description of a simple fact of the universe: things that are good at making copies of themselves become more prevalent. Hence things that become better at making copies of themselves prevail more than ones that don't.
Evolution has a very clear effective goal. Which it optimizes for, and meta-optimizes for.
The fact that this is a tautological drive, makes this effective goal even stronger than if it was by design. It doesn't drift. You can't change it. Even if you direct adaptations (i.e. replace natural selection with artificial selection), it still operates unswervingly. Only stopping if there is an extinction event, by happenstance or design.
Not only is survival relentlessly selected for directly, but anything that improves adaptability, repeatedly enabling more efficient downstream survival changes, is also selected for. I.e. meta-means to achieve this effective goal compound at multiple levels.
The highest meta-level, of course, being brains that know they want to survive. But that isn't some goal brains made up. It is an effective goal brains were invented for, and pre-imprinted with, as a means of better achieving it.
Anything goal-like in a living system can be explained at a low enough level by simple physics with "no goal". But if "goal" has any emergent meaning, which is to say, any meaning at all, evolution has a goal.
A tautologically emergent goal.
It isn't an overstatement to say that all other goals are either an expression of, or side effect of, that goal. I.e. curiosity, the need to avoid pain (even though statistically that sometimes motivates suicide), play (social and artifact behavioral exploration), and other seemingly flexible idiosyncratic goal generators only exist because of their statistical survival benefits.
So, THE tautologically emergent origin of, and unwavering statistical master over, all other goals.
Look, like Humpty Dumpty you can mean whatever you want with the word, but the unless you're talking about football or something the dictionary meaning is: "The object of one's ambition; a desired end or result."
You can't have the goal without the personality trying to achieve it.
This notion that evolution has an opinion is exactly why there are a lot of companion misconceptions about it - not least of which is the notion that humans are somehow at the top of the landscape it has produced.
Living things are relentless survival outcome machines.
Actively and adaptively. With every part and behavior continuously tuned, within and across generations, to support that one outcome. Carried out by inventive means and intricate strategies. With hierarchies of structure, function and interaction. And statistically successful over all kinds of environmental challenges and variation.
"Consistently, actively and adaptively, X outcome aligned", over all other potential outcomes, is as good a phrase for "goal of X" as any.
Yes, you can narrow the definition of "goal" in any way you want, for yourself. But my use of the term is consistent with normal use of the term, and its definition. It is not a technically defined term.
If you want to argue that there is a difference between "effective goals" for things that relentlessly and adaptively pursue some outcome, but without cognitive support, vs "reflective goals", created and/or carried out cognitively, often idiosyncratically to particular individuals, I would agree.
There is also a clear distinction between artifacts without structure and function aligned for some outcome, and those that "consistently, actively and adaptively" maximize the reliability of some very specific class of outcome. And calling the latter a "goal" is reasonable.
Hmm, perhaps I'm reading too much into this, but is this a side effect of the majority of the training data being from earlier years - much like I tend to vaguely think of "50 years ago" as being the 1950s if I'm not really paying attention...? :)
I believe so, see my result with Haiku extended thinking on. I think the weights are just too biased towards blurping out the majority of the training data of 'next year is xxx'. Interesting problem to solve indeed.
I'd be very surprised if the C/C++ code being written today was comparable in volume to the collective pile of JavaScript, Java, C#, Go, Python, and PHP.
Contrast to the heyday of C and C++ when not much else got a look in (Pascal perhaps? Perl probably...)
Sure, but I think that "back in the day your compiler would ..." is a reasonable characterisation.
Back in the day, yes, your compiler almost certainly would do that. Now, your compiler might possibly do that. For most working stiffs it won't though.
hmm how can I reuse this useful Go library in python... Oh I can't.. hmm and how can I reuse this useful java library in php ? Oh I can't. Oh and which of the programming languages you mentioned can and do use C libraries? All of them.
Reminds me of that coworker who thought that OpenCV was basically written in python.
I'm not claiming that there is no C or C++ out there. But it's such a nit pick when for most developers, no, their day-to-day work absolutely does not involve the creation of object files.
Sure, akshuwally, there are still C and C++ devs out there. Meanwhile a friend has just embarked upon a career as a pro COBOL developer. What of it?
Edit: Also, in the spirit of akshewally, I have just googled up this monster! My word, PHP and Java AND XML... it's like the unholy trinity of HackerNewsbane... https://php-java-bridge.sourceforge.net/pjb/
C++ is still wildly popular and will remain so for the foreseeable future. Your JavaScript interpreter/JIT compiler is written in it, so is your Java JVM, and I don't need to mention CPython in more detail.
Those all get compiled into object files and then linked.
The Player giving a bit of meta-commentary (meta-meta-commentary?) on plays in Rosencrantz & Guildenstern Are Dead:
"Well, we can do you blood and love without the rhetoric, and we can do you blood and rhetoric without the love, and we can do you all three concurrent or consecutive. But we can't give you love and rhetoric without the blood. Blood is compulsory. They're all blood, you see."
> Suspension would also allow the wheel to move upward relative to the frame and pedals when hitting a bump.
Perhaps I'm being dim, but I don't get why this would be bad; it just sounds like the definition of what suspension is!
Edit: I was briefly confused by the "relative ... to the pedals" bit here as well, but only 'cos I was thinking of the type where the pedals are directly attached to the wheel.
Location: Stockholm (remote or hybrid)
Remote: Yes
Willing to relocate: No
Technologies: Java, AWS, Linux, technical writing
Résumé/CV: https://www.linkedin.com/in/dcminter/
Email: dave@paperstack.com
I'm a British software developer now based in Stockholm, Sweden. I have years of experience building back-end stuff in Java, but I also have a good in depth knowledge of Linux, a good understanding of AWS cloud technologies, and a smattering of all sorts of other bits and bobs. I've mostly been working for Fintechs for the last few years - happy to do more of that or to turn my hand to something new! Smaller orgs preferred.
I'm personable, articulate, have a life-long love of technology, and I'm available with immediate effect. Let's talk.
Even more amazingly, I just discovered (from a friend after ranting about this on Facebook) that you're not allowed letters of this type in names on British Passports¹
Apparently there's a recommended transliteration table² from the ICAO that one should use to determine the replacement letter(s).
2025 feels like a very late date to still be conforming to the requirements of teleprinters. I wonder how long this kind of thing will be perpetuated?
I think parent poster had an X1 or something and assumed the conversation was about a similar contemporary device.
I'm a little sad this board isn't for my X220 ... I would be sorely tempted if it were - but like other posters I'd have some reservations about things like battery life even so.
By the (contemporary) by, a Mac Book is probably a better buy if you like Mac OS (I don't) because the hardware really is excellent. One physical point in favour of the modern Thinkpad though is weight - a MacBook Air is about 1.2 kg, whereas the X1 is not quite 1 kg.
reply