FilterJoe

data point: I started learning Python/Django 4 months ago (after a 20 year break from programmiung). Took me 40 minutes to solve. I'm not yet fluent with syntax so I had to look up (and still managed to mangle list processing syntax and was slow to interpret error messages related to accidentally swapping my variables for size of triangle and the list. My logic was fine once I got got the syntax down and swaps corrected. So your test would have caused me to be weeded out, I guess.

Had you asked me to pseudo code it I could have done it in less than 10 minutes.

Note, however, that learning web development requires learning many different pieces in addition to Python, so I simply haven't had the time to thoroughly learn all the syntax of any given piece.

Django is the leading Python framework (by any usage measure) for web development. Python is a popular language. Ergo . . .

As someone who has been learning Django/Python for 4 months, I very much appreciate both the getting started with Django site and the book 2 Scoops of Django that just came out. Would have saved me weeks of time if I had access to these 4 months ago. It's coincidence that they came out within a few days of each other (and a few days after Django 1.5 release candidate came out).

There's even greater interest in Ruby on Rails, the leading web development framework for Ruby.

I started learning Python/Django 4 months ago. At first I avoided deployment but when I got around to it, I easily spent over 1/3 of the last 4 months messing with most of the things outlined in the first lesson - and in fact I ended up making a VM (with VMware and dotcloud in my case) to ease setup pain for future projects. This one lesson alone would have saved me at least 2 weeks.

Drop the minutes and data plans to the minimum and drop your 2 regular phones. Then share your plan with other people if you can.

I am sharing a family plan with my inlaws at 550 minutes for the 3 of us. I also pay $15/month for data, so incrementally my 3rd cell phone adds $25/month + fees to the monthly costs. I keep cell data turned off most of the time except when I truly need to use it (typically Pandora or mapping while driving).

I typically use more than 550 minutes of phone time by myself each month as follows:

In office: Google Voice (free).

At home: My wife uses T-mobile@home service for $10/month. No longer available but you can use Google Voice comfortably at home with the right kind of phone router, such as this one: http://www.amazon.com/OBi202-Phone-Adapter-Router-2-Phone/dp...

Lots of high quality cell phone use on a single day: T-mobile $2 for the day (with my old Blackberry, which gives me much higher voice quality)

Our total monthly costs including fees and taxes are about $45. For that price we have unlimited domestic calling at home and in my office, I have an iPhone 4s with limited minutes and data and my wife has Virgin Mobile cell phone that she only uses for emergencies. Would only be $35/month if we got the above-referenced phone router and dropped T-mobile@home.

Feedback from a 4 month newbie: I've read several chapters. Overall the book is great but I have some specific suggestions, which I'll wrap in the context of what my biggest barriers to learning Django and web development have been.

Background: I starting learning web development with Django 4 months ago. I knew nothing but procedural programming (from late 80s) and a little CSS (Wordpress) before starting. I did Learn Python the Hard Way for a week then dove in.

Barriers to learning Django and Web dev:

1) Setting up the development and production environments has taken up at least 1/3 of my last 4 months of learning. This book would have helped with that significantly.

Suggestion 1: It would help even more if, in addition to your sample setup on github, there was a virtual machine or Vagrant file intended for beginners to get up and running quickly.

I've developed a VMware VM for myself along these lines (based on bitnami's Django stack on top of Ubuntu 12.04) and will use your book to improve it - especially changing my project layout template.

Suggestions 2: Answer a specific question the book didn't address: Is a virtual machine (one per project) a fine alternative to using virtualenv? If not, why not?

2) Django docs do a great job telling me what each command does, but rarely which of many alternatives is best to use (and why), and how best to structure projects. So I end up doing hours of searching each week (Google, Stack Overflow). This is where Two Scoops really nails it - would have shaved at least 2 weeks off my 4 months learning time. [can use this as a direct quote on your site, if you like]

Suggestion 3: Chapter 4 talks about keeping apps small and loosely coupled but not how to bind multiple reusable apps together, especially ones where tight coupling seems natural. I know there are techniques for this but I don't know which is best.

Suggestion 4: Chapter 6 briefly discusses when to drop down to SQL. Would love to see this expanded to several pages, with perhaps a long list of common situations where it does make sense to drop down to SQL. Either that, or a link to an excellent external resource.

3) Django is a tour de force in OOP. If you're weak on OOP, the tutorial and most other parts of Django is tough going. Practice and learning how to use my IDE (PyCharm) to navigate code eventually got me over this hurdle. This is not something addressed by your book, which is fine.

Suggestion 5: In Before You Begin section, perhaps add a single sentence that urges beginners to learn Python and (importantly) OOP before starting the Django tutorial.

4) No Django doc clearly explains the basic concept of web pages with no state, but context passed around to keep track of things. I still don't 100% understand why there are so many different ways of dealing with context, and which Django commands are best for given situations. Not currently in your book and maybe it doesn't fit in. But just want to mention that this was, and continues to be, a hurdle.

5) Deployment is way too hard for a beginner. I'm using dotcloud so that made it easier but again - having a virtual machine for distribution to beginners would be super helpful.

Suggestion 6: Expand chapter on deployment.

6) Would love to see an entire book devoted to building a single, substantial project that did a lot of typical things, but with all best practices implemented and explained in context. The project would use 2-3 existing Django apps and create 2-3 additional ones from scratch, and would show how to have intermediate apps to tie together 2 apps that need tight coupling. It would show this apps environments of dev, testing, staging, and deployment on either dotcloud, heroku, or gondor.io. It would show how to make the app look reasonably attractive, perhaps using Zurb or Twitter Bootstrap. I think this would make a terrific companion book to Two Scoops.

Hope some of this is helpful. Thanks for writing this book! I've had it for less than a day and I've already learned a lot.

Thanks for that - looks like a huge time saver! Unfortunately, it won't fully automate setup. Things like entering passwords for a service you want to set up, tinkering with desktop, firefox or other app settings, etc.

But still - this gets you a big head start.

roderikvdv's answer was helpful. I also found a long post discussing Django VMs specifically using vagrant:

http://stevelosh.com/blog/2011/06/django-advice/

Terrific concept. What tools (if any) exist to do this kind of thing on Linux VM? Do they work equally well for non Ruby environments (I'm thinking specifically Python/Django)?

I've spent the last week mucking around with VMware/bitnami to try to get a nice clean Python/Django/Pycharm VM for myself (with a template Django project that gracefully does both development and deployment). I'm almost done but what a long hard slog it's been.

I'm a different data point. I've used nothing but Windows (98, XP and now 7) for over a decade. Lately I've been learning web development (Django) and realized a few weeks ago that to really do this right I needed to do linux. So I used VMware/bitnami to set up a VM with linux/python/django.

It was Ubuntu 12.04 without the desktop. Very hard to get around with the command line so I installed the unity desktop.

Bingo. Very easy for someone used to Windows. I couple minor quirks but I didn't find that any of the above issues bothered me.

I have had issues. But these have been things like figuring out where to set environment variables, learning the bare command line basics, learning how to get postgres going locally, etc. Nothing to do with the desktop, except for the fact that it doesn't provide an interface for everything that can possibly be done with command line - so as (someone learning to be) a developer, I've had to learn all the command line stuff anyway.

I have no idea whether other desktops are easier. All I can say is that the command line had a steep learning curve, while unity desktop took just a few minutes to get started and I was off working.

Google has stopped doing anything with its interface but that hasn't stopped dozens of others who sync with Google Reader. I personally find that Newisfy (iOS) does everything I need for a Google Reader client. Don't like Newsify? Then choose from among many other apps on many platforms that sync with Google Reader.

I personally far prefer this arrangement to Twitter. I don't really like the Twitter concept to begin with but I can't even make it more palatable with a better client because they keep pulling the rug out from those who create better Twitter clients.

Learning Python the Hard Way was perfect for me. I did some procedural programming with Pascal and C in the 1980s, but nothing with programming for the next 20 years. 3 months ago I started learning Django/Python. Python was the easy part as I breezed through exercises 1 through 36 in less than 2 weeks. I agree though that exercise 37 is overwhelming and that's where I stopped learning Python and began building Django apps. I did also have to carefully study OOP.

I fervently wish there were something for Django along the lines of Learn Python the Hard Way. I HAVE been learning by doing projects and every few days I get stuck for hours on something confusing about Django I don't understand. I would love guided exercises - especially on some of the more confusing aspects of Django like the cumbersome HTML template syntax, HTML template inheritance, passing the context around, directory/file structure (and the kinds of issues to expect if your files are not in the right place), customizing forms, etc.

Of course, I understand there are different learning styles. Code Academy is even more pure learning exercises but I found that very frustrating as it keeps saying your correct output is done the wrong way but won't tell you why. Learn Python the Hard Way is in between because you have latitude to implement your own way and gives you creative license on some exercises to do extra if you're interested.

TLDR: Took less than 2 weeks to go from exercises 1 through 36. Taking me months to get up to speed on Django, which has no such exercises.

The first 2 lines (display, PPI) and the price line are why I did not buy an iPad mini, which would have been my first tablet had it been priced at $200 or maybe even $249. So marketing is dead on, in my view.

Instead I bought an Apple-refurbished iPad 3, for $379, which is $79 more than the soon-to-be-shipped Kindle HD 8.9".

For $50 more than the Mac Mini I get a bigger and better display. And for $79 more than the Kindle HD 8.9 I get better apps and user experience.

EDIT: clarified what first 2 lines were

For those who don't game or edit videos, this has been true for about 8 years. I retired my 2004 1GB RAM Dell desktop in 2012 and still keep it as a backup. The ever increasing speed of browsers and the shift of my work from the desktop to the cloud kept my system operating at the same speed it did 8 years ago - not blazing fast, but good enough.

I expect my Sandy Bridge replacement system to go another 7-10 years before I replace it.

I improved the post intro, it did get posted to lifehacker, and it was even introduced very well with "Joe Golton set out with $600-$700 to see if an ordinary person with no technical know-how can buy an off-the-shelf system that's also very quiet. Here's what he learned."

And yet dozens of people still felt compelled to point out how it is possible to do better with custom builds. I'm astonished at how out of touch the typical geek is with:

1) what a mainstream person is willing to do to purchase a PC.

2)There's a cost to all this geekery. You can buy a Lenovo e420s or a Mac Mini in a couple of minutes. Or you can spend a number of hours (5? 10? 20?) educating yourself on everything you need to know to build the absolute perfect custom PC for you.

I appreciate your compliment, which describes exactly my aims with writing style. It's actually much harder to write that way then stringing together a technical guide for the technical, which is what the first draft looks like for many of my posts. By far the hardest thing I ever wrote on my site was the extensive series on password management, which is a more technically challenging subject.

I'm really appreciating your posting it here as it caught the attention of Adam Pash at Lifehacker, who will be publishing it today or early next week. I made it clear with him that the audience is mainstream, not custom builders, so hopefully he includes an appropriate preface.

It's funny how some answers to tech questions are really very simple. My entire article can be boiled down to one sentence: If you want a quiet PC for mainstream purposes (no heavy gaming) for less than $700, simply get a Mac Mini or a laptop that has a mobile chip inside from the following chip familes . . .

I'm kind of glad you did because I realize the intro could be improved to make the scope and audience of the post clear (and acknowledge what else exists but is outside the scope of the post).

I'll improve the intro tomorrow.

EDIT: Added last sentence.

Agreed (I'm the OP). I consider HD3000 to be minimum acceptable graphics. Ivy Bridge has HD2500 in the desktop CPUs which is also (probably) acceptable and I could have stated that exception - but I was trying to keep the post simple for the mainstream audience for whom it was intended.

I wrote this article and was very surprised someone posted it on Hacker News, where the audience is highly technical. Your criticism makes it sound like the post was intended to be the ultimate guide to custom building silent PCs for the technically inclined. It was not. It was meant to answer the question: For about $600-$700 can an ordinary person with no technical know how buy an off-the-shelf system that is very quiet (and clean, while you're at it)? Among the thousands of possible off-the-shelf systems, what simple rules of thumb can you apply that insure the system will be quiet and clean?

Excluding my friends in computer/technology only a small number of people I know feel comfortable reinstalling Windows and hunting down all the drivers (I know how to do it and I personally find it to be an annoying waste of time). None of my nontechnical friends and acquaintances would buy a custom-built PC. I considered doing that when I purchased the systems and decided not to invest time in learning how to do it right (I agree with you that silentPCreview is a great resource for going down that path).

I considered the Mac Mini - which was quiet enough. But to run Windows Media Center I'd have to buy a copy of Windows and to run movies an external DVD which brings the cost up to about $1000. I considered boutique builders like Pugot but they all charge $1000 or more for their better Mac Mini like systems.

I actually started to write a much more detailed article half a year ago that you would no doubt have preferred but abandoned it because I thought only a sliver of systems building enthusiasts would care. My blog usually tries to make technical topics understandable by the masses.

I sometimes share my posts here but didn't consider sharing this post for a second - not the right audience.

I just joined Facebook a few months ago. I was afraid of it being a huge time waster with low signal to noise. So my expectations were quite low. Turns out the experience, with less than 30 Facebook friends, is even worse than I imagined it would be. The only thing I actually like about Facebook is the ability to reduce the amount of noise by blocking excessive posters and things like what video someone watched or what article they read. My main Facebook activity is managing noise.

There's a billion dollar business that's going to happen some day that will get you the information you really want, without the information you don't want.

You ask: Why not let them discover it for themselves?

In our social circles, many kids see Star Wars before age 5 and most have seen it by age 6. For many, that will be their first exposure to violence and evil. Depending on your values around violence (graphic or otherwise) and evil, you might want your kids to discuss and assimilate what they're seeing in a thoughtful manner.

Our kid had us read to him every part of the Star Wars story multiple times before we just let him watch #4 a week ago. It accomplished the main objective laid out in this nice article: Timing it so they're able to "discuss and analyze the themes appropriately."

I too feel movie #1 is one to skip, but for a different, simpler reason: It's about political maneuvering, which is very difficult to explain to kids. My 7 year old son still doesn't fully get it after reading the book several times and me attempting to explain in several different ways why Palpatine's strategy was so effective (cause wars in order to steadily increase his power). He needs a foundation of more basic political concepts before he can get these subtleties. My wife even had trouble understanding the movie when she first saw it. Without understanding what the story is about, it has the feel of random mayhem.

Back to the main point: There's indeed a lot of violence and our family was one of the last holdouts among peers. We knew many kids who had seen Star Wars before the age of 5, and we also know a lot of kids who think of killing and violence as trivial everyday things. Like the author of this post, we didn't want our kid to be brought up like that so we went through he effort of reading and thoroughly discussing the books.

edit: minor cleanup

According to a friend of mine who worked on the search team, Inktomi shifted its (management and CapEx) focus away from search and onto other projects. He thought at the time that even with the constraints of not competing with their own customers, there were things they could have done to better compete if management had chosen to do so.

Edit: fixed typo

Fair point but the 9700 is over 2 years old - and when it came out it had better battery life than the competition. The 9900 which recently came out has better specs and still a good battery life. What's killing Blackberry IMO is that it's nowhere near as good a pocket computer as iPhone or high end Android devices. The apps are much more difficult to install and upgrade as far too many reboots are required. And the difficult app development environment means there are far fewer apps.

I think consumers have spoken: "We love great pocket computers and we're willing to sacrifice voice quality, battery life, and data efficiency to get it."

The author asserts that "The first company—whether it’s an incumbent phone maker or Ph.D.-laden start-up in a garage—that figures out how to solve the smartphone battery problem will see enormous gains."

Nope. Blackberry has had battery life nailed for years and they're steadily losing market share in the U.S. I owned a Blackberry 9700 for a couple years and recently replaced it with an iPhone 4s. When using for a similar set of apps and voice calls, I get less than half the battery life I did with the Blackberry 9700.

I wrote a detailed comparison between these two devices (http://www.filterjoe.com/2011/12/23/blackberry-vs-iphone-4s-...) and there are actually a few ways Blackberry beats the iPhone 4s. The most important to me is voice call quality, where the difference is stark and impacts me every day.

Battery? I just keep it plugged in when I'm at home or at the office. Problem solved.

I tried it for a few minutes and it needs work. Adding a non-Wordpress.com site is awkward and results in getting all the posts and all the comments. I'm hoping Automattic rapidly improves it to become a real competitor to Google Reader.

The article and the conclusions of the underlying study bring up more questions than answers for me. Such as:

1) Does the importance of talent vary by domain?

2) They show working memory matters. Does working memory matter by different amounts in different domains?

3) What other "talent" factors work across different domains?

4) To what extent does having natural talent in an area lead to the development of poor work habits (because things come so easily) and thus long-term underperformance?

I don't have answers, but I believe that drawing firm conclusions about the talent/practice question is premature.

This makes for a great story but how do we know it wasn't totally made up? There is no link from the consumerist article to a post from the original source (someone whose purported first name is Ryan). Searching for word-for-word sentences from the article on Google, I just find this consumerist post and pages that quoted from it more recently.

Even if Laura Northrup was really given this information by someone who is really named Ryan, how do we know that Ryan isn't making this story up? Or perhaps there's a part of it that is true but important parts of the story are omitted or exaggerated?

This kind of thing seems to be happening more each year on the internet - widely circulated stories considered credible just because it's widely circulated.

All that being said - I own both a Kindle Touch and Nook Simple Touch but have spent a total of $0.99 on content, for the same reasons cited by many in this thread. However, I'm very comfortable with a Netflix model. I've used the Nook to read in stores a little and I intend to use the Kindle Touch with my prime account to read borrowed books.

I predict that some day, Amazon or Barnes and Noble will come out with some kind of "content guarantee: you buy it, it's yours, forever." And I think this guarantee will quickly be copied by all competitors. The market for e-content will then be extended to nervous Nellies like me.

Problem: Can no longer access old "notes" or "shared items"

Solution: Turn them into "starred items"

Details: You can only access notes or shared items from certain mobile or desktop clients. I don't know exactly which ones but the one I used was MobileRSS (iOS). Go into the "notes" folder and manually star every item (you don't need to star "shared" items because all of them are notes, but not all notes are shared items). If you want them in chronological order, you'll need to start with the oldest first, the second oldest second, etc. It took me less than 10 minutes to star my 131 items.

If you don't do this, you can't view or search for anything in "notes" or "shared items." However, you CAN view and CAN search them so long as they are starred items.

So far as I know, there is no longer a way to get content into Google Reader via the "Note in Reader" bookmarklet. It is no longer supported. This means that any site without an RSS feed cannot have any pages brought into Google Reader, so far as I can tell.

I wonder if there is a service that turns a site into an RSS feed on the fly? Then you'd be able to subscribe temporarily and star the item. Probably not convenient enough to use as a read it later (or archive in Google Reader) service, as I've been doing for the past couple years.

I've found that the easiest way to search for archive-worth web pages I've read years ago is to search my stash in Google Reader. Using a search engine doesn't always find old stuff which may be buried in search results or no longer on the web.

I restart the computer and chrome each morning. And when Chrome gets to slow I restart it mid day sometimes. I also clear caches once every few weeks which seems to speed it up a little, but not like it did when going from Chrome 14 to Chrome 15.

Anyone else noticing a big speed increase? Chrome 14 was actually much slower than the prior few versions on my 7 year old Windows XP system (1 GB RAM). I'm experiencing much faster page loads with Chrome 15.