I was engineer 12 at SendGrid and left after IPO and subsequent acquisition by Twilio. Being infrastructure and the backing many email marketing companies, we did really well. Kind of like selling shovels in the gold rush. We struggled more on the product front breaking into the much larger marketing space. Learned a lot there leading and scaling teams and scaling the email infrastructure to support over 8 billion daily sends.
Just remember that there can be hundreds of bottlenecks. Your second slowest matters too. And sometimes there are different requirements - the UI needs to respond within 1/10th second (sometimes much faster, sometimes much slower). Users can often live with your main calculation taking 10x longer than the optimized version so long as there is fast feedback that you are working. Eventually they will want the calculations faster too, but taking a minuter off of those is less valuable than a few ms off of your UI feedback.
you're still releasing resources - so you might not become faster overall but you can compute more in the same time later if necessity arises (althougth that might be somewhat premature but can be good for library code - so it becomes more applicable in different environments)
and there are some rare but tricky scenarios like:
hardware is mobile phone:
app seem to be bottlenecked on arithmetics according to the profiler, so it feels obvious to start optimization there
in reality what happens - hardware has limit on power, so it can't give full power to CPU, GPU and memory all at the same time
since app uses too much memory - it has to redirect power there, also memory emits heat, so CPU becomes throttled
By optimizing memory, everything runs colder -> CPU gets more power, can run sustained higher frequencies - app becomes faster
This one is more dangerous, as there may be backend resources in use that could be optimized, which could drop costs drastically. This may not change anything for your users, but is definitely worth exploring when it comes to getting the most out of your investment.
His boss is essentially making the same argument as Knuth: Spend your time optimizing what benefits the most from optimization. Or in other words, prioritize, don't optimize blindly.
It's a single line phrase, I wouldn't interpret it too literally. Usually you got to be pretty liberal when interpreting those easy to remember lines. It's a lot harder to remember the literally multiple books we could fill when bringing up exceptions.
The US constitution says congress will pay for useful arts and sciences. It says this before paying for national defense fwiw. If career soldiers and scientists can exist with federal dollars, so should useful artists. Now to define useful art...
The "useful arts" mentioned in the US constitution refers to the works of artisans and craftsmen, such as textile manufacturers, instrument* makers, and people working in construction.
*Realizing this might be confusing in context. I meant e.g. navigational instruments
My wife and I had our first at age 15. Then another at 22. And our last at 27. I've raised children while on welfare and while a software engineer.
I was more patient as a teen than I am now in my 40s. Now I am tired. All the time. I fear I would literally die of exhaustion if I had to maintain more irregular hours than I already do due to insomnia that I have developed over the last half decade.
Scale and costs. We are faced with logging scale at my work. A naive "push json into splunk" will cost us over $6M/year, but I can only get maybe 5-10% of that approved.
In the article, they talk about needing 8k cpu to process their json logs, but only 90 cpu afterward.
I know a guy who had something of an LSD overdose. Went psychotic and fully changed his personality. Total space cadet after. Ended up attacking an old man in a parking lot that he thought was a demon some months later.
This. There are certain events that we already know happened. We have approximate dates, and we have talked about the events with others, who have given us their impressions of how we acted at the time. But sometimes we don't remember what our internal thought processes were at the time, or we simply held some certain belief about what had occurred (e.g. "I just decided I didn't want to go to school some day", or "I wanted to go to school but ADHD wouldn't let me!"). Sure LSD can give us some creative visualizations, but sometimes it also gives us ideas that just make more sense than what we had before. Instead of some spooky brain disorder just randomly Stopping me from going to school that day, randomly, maybe it was actually a fight between alters (yay for dissociative identity disorder!). And the more I thought about it (especially being - apparently !! but clearly - able to see what was me and what was the other during those fights), the more true it seemed, especially given all the more instances of this I've discovered over time, and that it's part of a longstanding pattern, but I digress. More and more has come out over the months and years that builds a progressively clearer picture of why we act the way we do and why we've done certain things we've done; why I feel certain things I do, why I remember certain things I do and why I don't remember certain things I don't. I'm not seeing any universal truth; I am discovering our own truth, and we are discovering our own truths, and this has done a lot of good for us, especially where we gain a better understanding of our past behavior in friendships (and elsewhere), and where we can still improve, etc.
Even if everything we've learned through LSD were false, I wouldn't even really care because we've still uncovered and figured out so much through this (much of it we've discussed with various others who were there during past events) that we are incredibly grateful to now know. So overall I think the experience for us has been positive, that we have gained much from it, and that we may stand to gain more over time since we are a regular user.
None of that new information you provided is something that can be corroborated against the historical record. The claim is that LSD gave you access to long-locked up memories that happened to corroborate with what you are going through now.
The vastly simpler explanation is that your brain did what brains do best and filled in the gaps in your memory be inventing a story constrained to corroborate with the bare facts you remember, but because LSD it feels so real and vividly detailed that it “must be correct.”
> None of that new information you provided is something that can be corroborated against the historical record.
> The claim is that LSD gave you access to long-locked up memories that happened to corroborate with what you are going through now.
I think it's more like it gave us different perspectives on the same memories. We've had those memories but we couldn't see them that way before.
Also, uncovering this stuff randomly is what starts these paths of discovery for us. It's not that we're working through something and conveniently happen to find apparently useful things in past memories, it's that something (anything) triggers a flashback to a past memory that upon closer inspection reveals something that turns out to have had real apparent effects because it resulted in obvious behavioral patterns that multiple other friends have confirmed to us and are only now being placed in proper context with what feels like a proper explanation. This isn't somewhere the proper explanation could be just anything and this isn't schizophrenia where completely unconnected ideas are considered together. The stuff we find in flashbacks can completely surprise us and often has nothing to do with what we think we're looking for or think we're going to get out of it. It sometimes just somehow alludes to something far bigger that we really can corroborate with multiple external sources. Our brain and the others in it play so many tricks on us and on me, I know not to trust any single experience as a source of truth, but it can show me where to look.
> The vastly simpler explanation is that your brain did what brains do best and filled in the gaps in your memory be inventing a story constrained to corroborate with the bare facts you remember, but because LSD it feels so real and vividly detailed that it “must be correct.”
I don't really know how to properly articulate that I've verified anything because it's hard for me to even verify to myself that I know any of the things I've discovered for quite sure, and not just trusting something that very well could have been made up. I know the specific experience of recall is very fabricated, I know that what I see and what I think I identify as different parts while recalling those memories could be completely made up, but I'm not deriving everything from that experience. I use it as a suggestion to guide actual research that is not grounded in flashbacks and feelings and we have discovered actual patterns that are consistent with our theories and that couldn't possibly just be convenient explanations. Past a certain point all of Dissociative Identity Disorder is technically made up in some way (neuroplasticity!) and it's damn near impossible to know with certainty how things work or have worked, but I am taking the same approach that I have always taken and that is using the best explanations and the best theories that I currently have at my disposal and constantly testing and looking for improvements I can make to our understanding of ourselves.
But like I said, I wouldn't even care if it was merely a convenient filling of the gaps, because it has resulted in material good for us, it has resulted in material advancement of our understanding of our own past actions and in discussing them with those who we have hurt in the past, and it has also given us direction on how to better ourselves for the future. It really doesn't seem like that'd be the case unless the gap filling was so good that it correctly accounted for everything we hadn't even figured out yet, everything that we hadn't even heard of from others yet, and basically all of the ways we've been testing and testing the theories and the explanations that we have now.
Even if whatever it gave us access to was not something already stored in our brain, whatever it did was sufficient to allow us to figure things out that could otherwise have taken us months to years or even longer, and we know that they are material things that impact others because they have helped us work through emotions with others who formed those emotions independently of any LSD, based on those past actions of ours which we now better understand.
So I choose to believe we have gained valuable insight as a result of these experiences, not based on the experiences themselves which I know take great creative liberties, but based on everything we've done informed by the experiences to figure things out the old-fashioned way, too.
A common fallacy is that if you and I are smart then we will agree. Conflicts of interest and personal value systems and world views provide easy inlets for various counter points.
I used to know a guy that read some Hegel and then claimed dominion over Knowing the Truth and I used to piss him off by saying shit like “dogmatic people are slaves to their fears” and “discourse is a form of oppression”
Hegel is extremely satisfying on an emotional level to some readers. The idea that you —yes you, the reader— not only stand at the precipice of unfettered genius by the mere fact of possessing any critical faculty, but you also already possess the requisite tools to seize and contemplate the infinite truths of the universe, is very seductive.
Like everybody loves being told that they are smart, and Hegel tells the reader that they can be superlatively intelligent. It’s a hella thrilling idea to entertain, and the best way to keep that feeling going is by believing that it’s true, and the best way to convince yourself of that is to write defenses of Hegel.
reply