Sure it’s important to be aware of new stuff and able to identify which parts of your stack are being inefficient, but I do believe that changing the actual technology you’re built on constantly is not the best way to stay up to date. You’re better off just modernising the stack you’re on instead of constantly changing its base.
The guy with “Java architect” in his title might not be the best person to hire, but he might also be the guy who changes his old java code for new java code every time a major version comes out. He might use the builder pattern instead of the javabean pattern. Basically he’ll probably have a better understanding of how it works.
And this is coming from a develloper who is often trying the new thing, I simply mean that the available stacks isnt the only thing changing.
I do think the article is being a little bit polyannaish in its linear story of how new technology is picked up by progressive, nimble startups, who have good ideas and know what they’re doing, and then eventually is inevitably even if grudgingly accepted by the slow-moving, stodgy Enterprise. I see the point of this story, since it seems to be a blog by a group trying to push a “new stack” to enterprises, positioning it as the next wave that is already being used by their hip competitors and that they should make sure not to be on the tail end of. But it kind of skips over the new technologies that become hugely hyped and picked up by startups for a few years, and then disappear into the mists of time without ever catching on—sometimes because they turned out not really be good ideas. You don’t necessarily want to be on the leading edge of every hype wave.
Example: the stodgy, conservative enterprises that kept their data in an RDBMS instead of moving it all to MongoDB are nowadays starting to look not so dumb.
Yeah, my personal view is that the people who actually know what they’re doing move at least as slowly as the “Enterprise” (albeit with somewhat more awareness of why they’re being conservative). Don’t pick up something new unless it solves an actual problem you actually have, and you understand the downsides. But I suspect these “new stack” people would accuse me of being part of the problem and holding back innovation.
I wouldnt go as far as saying that they go as slowly, but yeah I agree.
eg.: My bank still uses windows xp on all their counsellors computers.
What these ‘new stacks’ are practicing is usually not innovation, it is re-invention in the flavor of the month. Incremental development is required for mainstream success.
The flip side of this: the industry moves at a fairly glacial intellectual pace! You can tune out most of the time and study papers without endangering your career. There are probably a few great, innovative ideas that emerge each year that you can catch up on as they are vetted.
Maybe I’m part of the problem, but how can you outsource your availability? It seems like at the end of the day, you own your availability. Amazon isn’t going to reimburse Netflix for the $$ that Netflix loses when an AWS region goes down.
Aren’t they? I imagine they’ll have a contract with an SLA and they’ll have negotiated what availability they expect and what the penalties are if AWS doesn’t provide it.
I can absolutely guarantee the penalties for an SLA violation are much less than the amount of revenue Netflix would lose over that period. I would be very surprised if they amounted to more than a free month of service.
And even within the SLA, when you’re having your .001% or whatever downtime, your ops personnel (who had better get paged, because your service provider is not the only possible source of downtime unless you think your “new stack” developers somehow write bug-free code…) will be up at 3:00 in the morning watching you lose money and unable to do anything about it.
Small point in the grand scheme, but I think it speaks to the mentality involved.
That’s not the single-page-app world that consumers increasingly prefer
I may be atypical, but I increasingly hate, hate, hate single page apps. A few examples:
When I go to Twitter, I have to remember not to click or scroll anything for ten seconds while the page components jump around and align themselves, lest something jump under the cursor as I click.
Whatever the fuck blogspot does that takes five seconds to load three paragraphs of text.
The ESPN mobile site has broken back. From a scoreboard you can view a game and navigate around, but trying to go back to the scoreboard results in a blank window.
ESPN also decided to go with the anti pattern of combining a bar full of links on the bottom of the page with infinite scroll. You can scroll down and see them, but unless you disable wifi you probably won’t be fast enough to click one.
Who exactly is it that prefers this shit?
I was going to edit, but since this is a somewhat new train of thought…
My gripe with this article (and its kin) is “change is good”. That’s no better than “change is bad”. Let’s just agree that change is different. Instead of advocating to do things differently, advocate for doing them better.
The table showing the evolution of development practices is interesting. In 2000, startups developed live in production. Apparently that’s bad, and they’ve evolved since then (while enterprise has remained still). But here’s the thing. Articles exactly like this one appeared in 2000 too, proclaiming the death of the slow enterprise. In another 15 years, will we see the same article with 2015 held up as the example to show how far we’ve come?
There’s the expression “everything old is new again”. Simplistically this means if you just wait it out long enough, you’ll be cutting edge again. I’m not certain that’s actually untrue however.
The article makes a point that one shouldn’t think of the stack as a capital investment. Perhaps this is why so many such stacks are shitty? (Citation needed.) Are we really arguing here that investing in ones dev stack is a bad idea?
I mean, I agree that Twitter’s website is pretty bad, but there’s a strong element of survivorship bias here. You only remember what’s broken, the JS-heavy pages that worked well went un-noticed.
That’s true, but I notice it on many, many sites. I do a substantial amount of browsing via a cell network, so figure minimum 200ms, possibly 400ms or more, latency. Everybody seems to be building sites along the gmail model, figuring I’m going to be spending the entire day with the tab open and so a heavy startup will amortize away. That would be true for gmail, but it’s not true for blogs or news sites or magazines or 95% of what I visit.
I’ve been confused a bit about this, but The New Stack, I guess is just a… blog? Its been sponsoring major conferences…
I poked around the site a bit before I submitted this, because it seemed like an odd piece… I half-expected it to end with, “So buy our thing!” I get the impression it is an industry group of non-competing hosting-related companies who got tired of hearing the same sales objections. As I’ve had thoughts along these lines and there wasn’t a product for sale or disconcerting ulterior motive, I shrugged and tossed the spaghetti against the wall – er, submitted.
Thats pretty much what I’ve gathered, too. The article was mostly fine. Thanks for submitting!
I find it interesting that a group like this would sponsor conferences, and send people to speak. Ricon 2015 has a speaker from The New Stack on the agenda. I guess I’ll ask about it.
Thanks for finding all these great articles. You’re virtually half of lobste.rs, hah!
Really? Over what time period? There was a specific time I decided to step up submissions, and I’ve not submitted much the last week…
Oh boy, another article venerating the brilliance of startups. In this case, their use of new tech stacks somehow means ‘smart.’
Look: they have their own fashion they’re slaves to. It’s still an orthodoxy; just a different flavor than that of the Enterprise. Theirs is a blind faith in inevitable, forward progress and the power of community to make bad ideas a bit less bad.
Oh, are you getting defensive? Sounds like you’ve been too busy developing and not busy enough trying new things over the past five years
This statement alone could be used to justify anything.