@P.Fonseca I honestly have no idea what he means when he claims that the PDF generation failed miserably... This is likely on the low end for people here, but I've written upwards of 50 pages of PDFs in Mathematica, full of nicely typeset math, graphics, and code, and never really had trouble (although I do vaguely remember having Export crash my FE once or twice, while Save as... worked beautifully)
@P.Fonseca @P.Fonseca Yeah, very interesting. It didn't take long for a reaction to The Atlantic article with a special spin on the social aspects between the development and usage of the two systems Mathematica and Jupyter. Discussion continues also under the YCombinator.
As @b3m2a1 I do not get the technical complaints of P. Romer regarding the static publication via PDF. I'm trying to not get too upset about it. It's quite bitter that some folks can't even appreciate the WL as it is and just like MMA output and number of integrals MMA can solve.
As I mentioned previously, there’s something that needs to be done in respect to joining the “open” community. The cloud notebooks have (and are) played/ing a role on this openness. But current scenario requires for the language to be available on other environments, for people to perceive openness. One thing that is hugely forgotten on this “social” debates is the the fact that we are all paying for these free tools, either we want it or not, either they are good or not, either their...
... production is efficient or not (cost per benefit); as a lot is this money is coming from our taxes, or other products that we use directly, or indirectly (Microsoft, Google, etc...).
If we extend this free is the way notion to all other fields, will I eventually receive an iPhone for free? as I would have payed it with a little bit of my taxes and a little bit of my unconscious glancing at Google adds?
On the open aspect, well, if the notion of value and proprietary was absolute (that “free” is kind of participating on its weakening; and I don’t believe there’s too much debate on this, as the “free” internet has brought a “free” problem to a lot of copyright industries) a lot more would be opened (because, lets be honest, many want it open for the noble reasons of science, while many want it open for the envy nature of human condition...)
But again, a balance is needed on our beloved working environment... Full openness is hard as it would be difficult for WRI to strick a 1.0 billion $ founding contract, that would allow them some years of financial revenue, to adapt to the new business model, while keeping some management and direction independence (a value WRI cares a lot about, and with some merit at their 30th anniversary...), during which, for sure, a substantial percentage of proprietary content would be stol... adapted...
...and gracefully adopted in other competition environments (because it is not because things are open and free, that there’s not competition or victims... as most free production does get’s paired...).
As I have mentioned before, this extra step could be the micro kernel, preferably written in WL itself, and compiled with the new compiler. Add to this a few “libraries” that are currently not at the top notch when compared to what exists for free, plus a couple of gems that are top notch, and some interest from the open community might appear... but there’s some gambling... although, currently, the status quo is also a big and risky gambling...
(paired /. paired -> paid)
(And sorry for othe misspellings, as a tablet is not the greatest interface, most specially for my generat... past experience...)
Is there any built-in Part/Extract like function that takes a default for non-existing parts? (Similar to Lookup, but not for associations) - I can't seem to find one...
A found a notebook corruption bug in M11.3 (haven't tested other versions). Something the notebook source is formatted in such a way that a line break is inserted in the middle of an -> operator, breaking it into - and >, and thus making the notebook unreadable.
@Kuba D'oh. Over the last few years I've come to think I have dyslexia. I don't know whether I'm becoming more sensitive to something that was already there or it's getting worse. I'm pretty sure I flipped the gravatars, but I usually click the reply button....Not this time, though.
@MichaelE2 thanks, something like that should work - I guess you could even do something like ReleaseHold[Check[Extract[{a, b, c, d}, {5}, Hold], Hold@Missing[$MessageList]]]` if you want to be very sure that the message is coming from the Part and not something else ;) just seems a bit strange that there's no built-in for this
Replace is probably the easiest to control (in the use case I'm imagining), if you want to turn it into a lookup-type function.
The Block one zaps all instances of Part that get evaluated when passed into the Block. Potentially buggy, though I can't imagine such a situation normally arises.
@b3m2a1 he is too harsh about pdf issues but he has a point. "pdf export looks ugly" questions appear with constant frequency. very often issues are os dependent, and many time require setting explicit value for an ambiguous style related option. For a power user like you it is a little annoyance or you may even not notice it by using 'proper' os etc. For a newbie who only knows more or less how to work with notebooks, it is couple of hours wasted on finding solution. I've been there too.
@MichaelE2 mistakes like this happen for me too, because I'm not as good as in the past in handling concurrent 'threads' and I mix them :) So, was there something else you wanted to tell me? :)
@Kuba @b3m2a1 I too believe that his argumentation is correct. It's not about that the PDF export works or looks ugly. It is about that SW and the Mathematica Journal obviously uses something that makes quality output. If you just use a notebook with mixed content like this, I'm not sure you would get the same quality.
