Can you explain? What I’m reading here is people discussing drawbacks of various non-string-based systems, which seems like an appropriate reaction to a guy telling them that maybe people use strings because the non-string-based stuff sucks. (Not being well-known or available in a widely-used language is a drawback in this context!)

And the author made the (annoying) point that I'm now ready to call as a bit of an old-timer here. When people keep recreating the "bad design pattern" over and over, you should probably get over it and roll with it.

> [Y]ou should probably get over it and roll with it.

I’m not sure that’s what he’s saying:

> If people want to displace string templating, figuring out what those current advantages are and how to duplicate them in alternatives seems likely to be important.

But that’s not an interesting objection—if you want to say it, we might as well use that to justify talking about it instead. What is interesting from my point of view is that I can’t see what it would actually mean to “roll with it”.

Stop trying to invent something better? Thanks but no thanks. (I’m just a sucker for potentially extremely neat things with a long history of mostly failing—structural editing, live programming, graphical programming... I doubt anybody can reform me at this point.)

Try to mitigate problems that result from this? If there ever was something that failed even heavier and in even stupider ways than eliminating string templating, it’s web application firewalls and their ilk. At least I haven’t ever heard of them stopping a determined or even somewhat competent attacker.

Try to trick people by doing something that looks like string templating but is in fact syntactic? Worth exploring, but doesn’t really count as rolling with it, I think.

The only thing I can imagine here is tainted strings, and those do work, but like the previous option they are hardly seamless. Something else? What?

> I’m just a sucker for potentially extremely neat things with a long history of mostly failing—structural editing, live programming, graphical programming... I doubt anybody can reform me at this point.

There exists a cohort of people, so called “harbingers of failure”, that inexplicably prefer and buy new products which turn out to be flops. I suspect I am one, too.

The topic of this strange kind of people seems to be discussed here quite a lot: https://hn.algolia.com/?q=harbingers+of+failure

You could probably be one of such people. I think you should document your preferences somewhere public, so that we know what else is likely to turn out to be a flop.

> tainted strings

At least in Perl’s implementation (one that is famous among me) it’s possible to untaint them accidentally by doing some innocuous operations which may not be directly related to their final purpose.

I think you still need to explain further. it is not clear why you think the author is being annoying. it sounds like you are agreeing with the author in your last sentence. I'm confused.

No matter what other damage it causes?

"Roll with it" probably means that we should expect people to use strings for templating, so we should design other downstream mechanisms to handle or prevent potential damage.