Lanny Gilbert

Thanks Ulf! My thought exactly. Building in GraphQL level functionality would make me look cool   but given that the client list and the functionality isn't going to change, why go to that trouble? Just do "regular ReST" and be done with it.

Totally agree with your post Tim. My question was more around "Given that I *HAVE* to migrate these services, is it extra effort - and is it worth that effort - to use GraphQL vs. ReST".
Methinks the answer, given our clients' reluctance to do anything , is to build ReSTful interfaces that require less effort on both sides, but wanted to see what the community had to say.

I'm working on a project now where we're taking 15 to 20-year-old code (some CORBA(!!) interfaces, some SOAP interfaces) and attempting to drag the code (and our clients) kicking and screaming   into the 21st century.
My questions for Samer Buna (and anyone else, especially those who are facing or have faced something similar):
   - How much more effort will it be to implement a GraphQL version of the server-side code vs. ReST? Or is the effort the same (more or less)?
   - Our clients, for the most part, want to do as little as possible in order to comply with the change from CORBA and/or SOAP to ReST or GraphQL.
     Given that these APIs have been around for 15-20 years, is it worth building GraphQL, with all its flexibility, instead of ReST, given that the clients will be sending and receiving the same data points, only in JSON format rather than wrapped up in CORBA or SOAP cruft?
     Also, these services won't be exposed to any other clients (although they could be, but the current plan is that they won't).

Thanks!

Thanks for the reply, Daniel! Looks like this book is right up my alley. "Throwing fancy stuff around" seems to be the mantra of many computing books these days. Nice to see someone starting slow and building up, using
time-tested computer science techniques, to the best solution. NICE!

Not as such...  

You're right! I was just mentioning that we talked a lot about sorting *and* a lot about O(n) vs. O(log n), not that - as a 2nd year undergrad - I could write a sort in anything less than N(log n).  

When I was in college (at Georgia Tech - Go Jackets!) the language of choice for all but one of my CS classes was Pascal. Yes, I'm old!
I remember one professor, who seemed irritated that he was required to teach lowly undergrads, going over a fairly complex algorithm for a homework
problem in class. At the very end, he said "Well, you can't do this step - he pointed to a line - in Pascal, so I'm not sure what you're going to do". That's my funny story.

Now a question. GT required 2 classes of computing theory, purely on algorithms and also how to make algorithms more efficient (a lot of time on sorting) so they'd
compute in O(n) vs. O(log n). Does the Algorithmic Thinking book go into this kind of detail? Also, perhaps the pros/cons of B-trees vs. B+-trees?

Just the title of this book brings back many memories of a lot of late nights.

Thanks!

Hi. I looked at Manning's page for this book and it looks VERY interesting. I have 2 quick questions.
Do you go discuss Jupyter notebooks at all? All of the data analytic work I've done so far used Jupyter as a platform for the Python work.
Do you talk about tips and tricks to make "vanilla Python" highly performant, or does the book concentrate on Cython for the big-time performance enhancements?
Thanks!

I am being charged with setting up a runtime for a vendor product that requires a (in my mind anyway) pretty large Cassandra instance to operate properly. Does your book cover how to use Cassandra (Cosmos) within Azure, with regards to sizing, monitoring, best practices (over and above standard Cassandra best practices), etc.? Thanks!

Claude, that raises a good point. If I learn JS in order to access TensorFlow libraries, I’ve started to get a good JS foundation so I can use other powerful libraries like Angular for front-end and Node.js for server side. I’m not aware of anything similar for Python. Also, there are libraries in JS that allow you to write in JS and then compile into iOS or Android native. So seems that learning JS to use TensorFlow is a gateway to all kinds of other technologies while Python doesn’t seem to have that. At least not that I can find.  

This question is for Shanqing, Stan & Eric specifically, but everyone in general

This question is for Dane specifically, and everyone else in general  

Python is GREAT for doing Machine Learning, due to many factors, the number of supporting libraries available being foremost to me. Other than that, why would I want to use Python/Jython over Java, given that I’ve been using Java for 20 years. What class of problem(s) would Python/Jython be a better choice to use?

Thanks!

Hi. All the developers I work with use Python exclusively for ML work. I have dipped my toe into the water in the ML area and found Python an excellent tool (even though I had to learn it) with many libraries that make building ML code very straightforward.
JavaScript, on the other hand, doesn’t seem (to me anyway) to be as clean a way to do ML.

Convince me to switch my way of thinking and thanks for taking my question.

Null safety?? I know one manager who’d drop everything and send his team to Kotlin training just to get rid of the daily train of NPEs from the Java code in the app he manages. Thanks for the reply. Gotta try this out 😎