It's both.
The Goddamned Consultants are the worst with the cargo cult crap, and we've got someone high-up egging them on...
Me, I'd be a lot happier if people would stop arguing over what is the best framework du jour, stop neglecting a working stack just because it isn't trendy, roll up their sleeves and start taking some responsibility.
Peh.
Consultants are always going to be the worst of the worst. By being fad-compliant they can present themselves to management as being cutting-edge while billing way more hours than otherwise would have been required to complete a project.
I actually like JavaScript as a language but it's true that the whole node.js thing got way too big too fast.
I'm not a fan of Perl, as several folks here know. I prefer my code to look legible.
(But then why, you may wonder, would I prefer to code in C++ if that's true?
It's complicated).
When using a scripting language, I'm fond of Python.
Some of the Rails programming I got into recently was pleasant enough.
JavaScript? Eh... I could take it or leave it. I don't have a problem with the language itself, per se, but I could see if you had to work with it across different browser environments where those different environments might make it annoying.
Java? That's a language encouraging a sort of broken mechanic with more broken-ness on top of a broken idea. It's so seductive, especially when you have things like Eclipse that makes it almost a visual language with its point-and-clickness (and, seriously, I do like that feature in the IDE), but when you have to do something large with it, and everything looks like objects just for fucking objects' sake, you want to just get something the fuck done, but it actively gets in the way, insisting you try to think like it, rather than like a human being.
cutting-edge while billing way more hours than otherwise would have
been required to complete a project.
Ding ding, winner winner chicken dinner.
Their architectural approach is "build a big huge webservice module deployed to a separate machine" for a small bugfix that should have involved writing a new .class or two and wiring them into the existing execution flow.
"Job security", I guess.
Mon Jun 12 2017 03:27:35 PM EDT from LoanShark @ Uncensoredcutting-edge while billing way more hours than otherwise would have
been required to complete a project.
Ding ding, winner winner chicken dinner.
Their architectural approach is "build a big huge webservice module deployed to a separate machine" for a small bugfix that should have involved writing a new .class or two and wiring them into the existing execution flow.
"Job security", I guess.
Not sure what consultants you guys use....or maybe coder consultants live in a different world. My InfoSec professional services consultants could never get away with that! We constantly are forced to justify every approach and plan - and hours are watched like 4 hawks in surround sound. It's often hard just to get clients to pay for the most basic of things they need to avoid breach, much less "nice to haves"....forget about extra bells and whistles.
These guys are known for hopping from start-up to start-up, each time working with relatively recent (or inventing entirely new) technologies because it's the latest, greatest thing. They've never read Mythical Man-Month, and so are completely unaware that a silver bullet doesn't exist; or, they simply don't believe it, and are hellbent on trying to invent one anyway.
The result is a pile of mostly identical feature sets in otherwise wildly incompatible tools or languages, requiring otherwise responsible engineers to become generalists and forcing them to learn whole *classes* of technologies instead of settling down and focusing on a small set of orthogonal and portable tools. They decry vociferously how one should not reinvent the wheel, hypocritically ignoring that they did exactly that months to years ago when they spawned their niche of the open source community. (It also ignores the fact that I cannot remember when I last observed wagon wheels on a Mazda RX-7, but I digress; that's a rant for another time.)
Meanwhile, after they've bounced from their current place of employment, these serial entrepreneurs leave the husk of their flagship projects for the rest of the "responsible engineers" to maintain in their absence, devoid of documentation, crushed under the weight of schedules now coming due, and with the routine conflagration of operations infrastructure one tends to get with underspecified, immature technologies applied at scale to real-world problems.
My previous employment actually took measurements: 75% of our 12-man team's time was spent actively dealing with ops-related issues; only 20% remained for productive work. (The remaining 5% were spent in meetings and the like.)
Probably not a bad rule of thumb, although it sounds a trifle simplistic.
I don't know if I'd want to use either for web development.
2017-06-13 12:24 from fleeb @uncnsrd
Probably not a bad rule of thumb, although it sounds a trifle
simplistic.
I'm sure a highly paid consultant said it.