3 ms·
The Wayback Machine begs to differ: https://web.archive.org/web/19981202230410/http://www.google.com/ https://web.archive.org/web/19981202230410/http://www.goo
by kkielhofner 4y ago
The Wayback Machine begs to differ:
https://web.archive.org/web/19981202230410/http://www.google.com/ https://web.archive.org/web/19981202230410/http://www.google...
I'm clearly not talking about Google as the behemoth we know today - I'm talking about Google the search engine (yes, Page Rank) - which I was using in 1998 because it was already vastly superior to Altavista, Lycos, etc.
I was on dial-up until 2004 (DSL, yeah!) but I was skipping trips to the library for homework by 1998 (at the latest) because the web and the fundamentals powering it (routers, modem banks, operating systems, server and client software, etc) were already vastly more mature than what we see in blockchain today (see my parent comment). HTTP 1.1 was published in 1999 and it still works today. Yet here we are, in 2022, with the entire Ethereum network being given a couple of weeks (at best) to deploy a massive software upgrade for a fundamental protocol level change or get left behind. Sure enough - 88% of clients aren't ready for it[0].
I don't know about you but I don't remember a single instance of "if you don't upgrade your browser in the next two weeks it will not work at all" in my three decades of being on the internet.
I was in a rural area and could only get 14.4 on my modem reliably (often 9600). It was slow but it worked. Besides, are we really comparing blockchain software reliability and quality in 2022 with the millions of miles and countless components of physical infrastructure that it took to bring the web to hundreds of millions of people by 2000?
The web was slow (like blockchains aren't?) but still worked. It was slow because it turns out literally digging up entire countries to deploy broadband and laying submarine cables around the world is (to say the least) a challenging, extremely expensive, and long process.
[0] - https://ethernodes.org/merge https://ethernodes.org/merge
- chrisco255 4y ago> The Wayback Machine begs to differ: Yeah, I get that the website google.com was around in 1998. I was around then, too, and used it too. But behind the scenes of that deceptively simple looking frontend, is a mass of data centers and a multi-million line code base that was (and indeed, still is) being updated and maintained by thousands of engineers to prevent it from falling apart or being abused by SEO maximizers to the point of being unusable. If it's so simple, please try to re-create it. It shouldn't take you too long. Just a wave of your hand. > if you don't upgrade your browser in the next two weeks it will not work at all" in my three decades of being on the internet. Plenty of back-end servers with zero day exploits have to be patched all the time. Other incidents happen all the time behind the scenes that engineers are forced to fix on short notice to keep a massive service operational. I'm guessing you've never worked on call as a dev ops engineer. Ethereum works. It has 100% uptime since 2015. That's impressive. Not even Google can claim that.
- kkielhofner 4y agoI have no idea how we're still talking past each other but assuming good faith I can only think we're having a basic communication problem? I'll try to be more explicit. Are you claiming Google had millions of lines of code and thousands of engineers in 1998? It certainly didn't - as best I can tell from history it had less than five people in the entire company at that point. My point is Google went from research paper to you and I using it in two years with a few hundred thousand dollars. For the record - I couldn't do this today let alone then and I don't know where you got the implication I said I could? Yes, over 25 years later it does roughly six billion searches per day and is what you describe. 500 hours of content are uploaded per minute to YouTube alone. I also didn't say Google was simple - actually the opposite. Again, point is a tiny number of people with an accompanying tiny amount of investment (relative to almost anything in blockchain today) did something very hard and had it in the hands of you and I in two years. Back to my original point, Ethereum is seven years in with an army of people (808 contributors to geth alone) and billions of dollars. Yet, as this merge has shown so far the implementation is fundamentally still a mess. I'm trying to make some comparisons between the early (10 or so) years of the web and blockchain. We don't have 30 years of blockchain to look back on so I have no idea why you insist on bringing the state of Google in 2022 to the discussion as it has no relevance whatsoever.