I'm serious in the sense that some of the extreme syntax-sugar and leniency, more than one way to do it, etc, that Perl is loved for are the reasons that people dislike it in a modern context, and these are to simply be removed with no equivalent replacement.
If you move toward a readable standard "Perl" that supports (only) modern coding styles, you will have something that old Perl users wouldn't really recognize as Perl.
Consider something as simple as:
https://news.ycombinator.com/item?id=4774426
This is "Perl" and recognized as such... but my suggestion of calling some entirely different language Perl in the future would make this essentially a non-script. The same is true for most syntax, as well as a large amount of fundamental structures, regexes, etc.
I'll speak following my analogy: I'll call Perl the Apple ][ and some language that wouldn't be recognized as Perl by today's Perl users, but which should be called Perl in the future I will call the iPad.
For example, it's fairly clear that the iPad language should have every single variable be a full object. The Apple ][ does not force this. Likeweise, classes and OOP would not be optional on an iPad, but a fundamental part of writing any program. But on the Apple ][ it is optional. On the iPad it is clear that $_ would be meaningless and not used, ever. It's a syntax error.
On the iPad all source code is UTF-8 or Unicode, and all built-in standard operators including the iPad version of a regular expression (which uses almost none of the syntax of the Apple ][ language natively operate on UTF-8 or Unicode.
The iPad language is longer and has more explicit, longer keywords. The syntax itself is extremely orthogonal and you could describe the entirety of it to anyone who has programmed in any language in a relatively low number of slides, and they can just start coding. It is fully readable to someone who has never read the iPad language, i.e. if you were to read a numerical algorithm or datamining algorithm that was just talked about in a textbook, you wouldn't be distracted by the textbook having to explain what the iPad is. It speaks for itself.
This explicitness is part of the language; there is no way to write one-liners or extremely concise hacks the way the Apple ][ could.
The iPad language lives on the web. It has primitives for handling a client connection as well as a standard framework for rich client-side connections. A persistent connection over http is as everyday as standard input and standard output was on the Apple ][.
And boy can you Google this. The Apple ][ could have a syntactic structure, special variables [1] that were almost impossible to Google. All sorts of Context.
The iPad language has no such context. There is very little you can set to change the fundamental syntax, no way to change the default index on arrays and related things, or any such high-context things or syntactic context. There is just one context, and everything that is done in a program is spelled-out and eminently Google-able. You never have to ask what something actually does, on a syntactic level, or figure it out by running it. You don't have to spend even six days to get to this level at the language: an afternoon tells you all the context that you will ever need. Anything beyond that, you can Google.
Sure, you may not understand what a piece of code does on a higher level. What happens if you apply a fourier transform to _____. It may not be obvious.
But syntactically, it is extremely easy to read off what you're doing. No more difficult than reading off pseudocode in a textbook.
And the language is succinct. The basic syntax is so small, so orthogonal, that you can be handed a 20-page description, having never seen an iPad language, and be able to more or less implement a version in the language of your choice, in a few weeks, without any further information or context. There is not a lot of magic involved.
Of course, the only thing that the iPad language will have in common with the Apple ][ is just the brand (Perl)..or the fact that at some level they're both turing-complete.
The future of Perl is calling an iPad language that is unrecognizable to any current user of Perl, Perl.
Gone are the days of syntactic sugar or short little idioms.
[1]http://www.kichwa.com/quik_ref/spec_variables.html