JavaScript, holds PDFs hostage, adds zero value (my browser can display PDFs just fine, thank you very much), operates a business based on copyright violation.
The last time I've checked, the worst thing was that it demands money once the document is old enough(!) Then, it required from you to have an account just to download the document. Finally, it didn't work without Javascript. For me, instead of adding value, it reduces it, so I decided to avoid it to save my time.
Not just Javascript but Javascript from about a dozen different hosts last time I tried it and quite a few of them seemed to be required. I avoid it too.
For a platform that is supposed to enable the user to read documents online, it fails by not enabling the user to, you know, read documents online. Oh, sure, sometimes I might get lucky and be able to read the first 4 of 37 slides without the application stopping to respond or crashing my tab.
It's still very slow. They have improved in this regard, but the original performance was so bad that saying that is not a compliment.
There's a pdf viewer built in to firefox now, which in my opinion is superior.
With noscript, the text is not formatted correctly.
With noscript off there's a banner across the screen, "download full view" etc.
This banner is rendered correctly in chromium but not my version of iceweasel, where it is about an inch down from the address bar and obscures the text.
Pages after the first one don't appear until some time after scrolling down.
Finally scribd want me to sign in to download a pdf.
I'm not sure what you mean by "stumble across documents I'm not supposed to read".
It used to have a really obnoxious and entirely useless UI around the actual document. But recently links to scribd point to plain PDF so I made my peace with it.
Isn't this done automatically with all submitted PDFs? At least I seem to recall that there is some integration between HN and Scribd (YC '06 incidentally, which might explain it), but there's nothing in the FAQ or anything.
This is why I still like old Unix systems (early BSD/v7 etc). It is possible to get a complete understanding of the system in a relatively short amount of time and they're still pretty productive.
Reading the link felt like watching a movie that gets interesting and stops midway. I wanted a bit more, i hope the author expands on the article & turns it into a longform one.
I am slightly disappointed by the list - for example C# threading is a book from 2003 and .NET has evolved a lot since then (Task, async, etc.). There are other, more recent books that represent current state of art in C#. Perhaps the reading list shouldnt state the year for which it is intended for?
I'm reading Programming in Scala. The free online, out of date but good enough to help you decide to go further or not first edition. I have the scala repl open alongside. Never programmed in Scala, don't know that I ever will, but it's interesting. http://www.artima.com/pins1ed/
I'm also re-reading Niven's Ringworld series, as a bridgehead to the rest of his Known Space material; I've never gone beyond Ringworld before. I'm in the third book at the moment. https://en.wikipedia.org/wiki/Known_Space
Besides the update on the official book, Bjarne has written a new one that teaches how to write modern and secure C++, instead of using the unsafe C underpinnings.
Planning to switch from c to c++; I'm really exited to read this.
@commenter: The word switch doesn't have to rely on relevance between subjects it is referring to. It has more to do with the speed of changing the direction.
i suggest you phrase it 'planning to learn c++' - there's hardly any switching from C if you plan to write modern C++, because there's hardly any C in it.
That has been my problem with C++. When I took my classes that used it[1], the ink was still wet on the original standard. After that, I spent years writing real-time C, with the occasional foray into "C with classes" territory. I can't seem to shake that foundation and "learn" C++ again.
[1] I'm an EE/borderline CompE so I only had three.
I am certain you will find any number of lists, narrow and broad, on Reddit, GoodReads, LibraryThing, O'Reilly (blog posts), Amazon listmania, StackOverflow, HN etc.
There are some good books in there for embedded/firmware engineers. "Making Embedded Systems: Design Patterns for Great Software" and "Multi-Core Embedded Systems" look interesting.
Hennessy and Patterson is in its 5th edition. That makes me feel old. I can only vouch for earlier editions, but this is a good book.
(I'm only familiar with the 1st and 2nd editions of this book, and of the two, I greatly prefer the 1st. The 2nd has a lot more information (and of course is somewhat more up-to-date), but that only seems to hide the more significant information. More complete is not always better.)
Am I to presume that this list carries as much validity as most other "recommended reading" lists?
/which is to say, books which most working in the field didn't read to get into the field, don't read during the actual job, and probably aren't going to read except maybe as a reference chapter here or there?
Yes. I looked at the software dev list; I only skimmed it, but none that I saw were books I read, or books I'd care to read, despite being in the field for 4 years now, and being a fairly passionate learner in terms of new techniques, technologies, languages, and methodologies.
With only four years' experience I'd expect you to know little about any of the fields on this list. I'm surprised you found nothing of interest, unless you are only focused on web stuff.
I assume these are intended specifically for Intel employees (i.e., targeting technologies and areas they have internally), rather than for any positions located elsewhere. And there, possibly meant more as a reference than a teaching tool.
A surprisingly low-quality, diluted, conformist list of mostly boring and outdated books. You can't expect much innovation to come from a team reading this. I scanned through the list 3 times looking for something of value, but nothing caught my eye. Disappointed by Intel.
Well, Java Concurrency in Practice certainly isn't low quality but it is boring and outdated. An alternative would be Clojure For the Brave And True :)
Java may be outdated, but "Java Concurrency in Practice" is not outdated if you're one of the millions of Java programmers still working in that language.
The point is to get the reader ahead of the curve! It's not really Java they're suggesting you learn: it's concurrency. Once you learn concurrency in one language, much of your understanding carries forward to other languages.
Intel has told us multicore is the future. They're trying to get you ready for it. Less certain is whether Java or Clojure will be the new cool thing.
Only Android Book I would get is "The Busy Coder's Guide to Android Development[1]" as it comes with continuous updates. It's also by one of the most knowledgeable app developers in the Android Community. It's worth it alone for how much he cares about backporting within the book and not just supporting the latest versions of Android.
Android development has not drastically changed since ICS, but there's enough that changes still that I would be reluctant to waste money on a book that does not update regularly.
See, that's part of the problem. It's not at all clear what this list aimed to accomplish, which explains the low quality of the result. You want me to recommend a book? Sure. Read Hemingway.
Reminds me of my response when someone on the street with a clipboard says "Can I give you some literature?" I'll sometimes answer "Sure, do you have any Steinbeck?"
It turns out it's a pdf served at this address, along with another link to scribd in the brackets:
https://noggin.intel.com/sites/default/files/Intel-Recommend...
"I think the title should read [pdf] and [scribd]"