Check the ISP's specs for what kind of modems you need, and get one that matches.

That's what I did, to go buy one at Best Buy instead of renting one forever and paying way more in the long run.

Really depends on how your ISP is setup. If they support docsis 3.0 then you should definitely get a docsis 3.0 modem, both the Arris and Motorola are good choices.

I don't know the exact details of how it works. It lets your connection bind to more channels which in return could potentially give you more speed if your ISP overloaded the nodes in your area.

In my case my ISP is advertised as 15mb/2mb for their basic plan. For years I was getting 8mb/2mb and it never went faster. It also dropped to 1-3mb/2mb very often during prime time and sometimes even during the day. After asking them to switch to a D3 modem I started to get 23mb/2mb speeds all the time.

It's not a magic bullet though. You need good signals to get reasonable speed too. Signal levels could deteriorate over time. It might even just be that a splitter went bad, that could give you abnormally low speeds 24/7. Your ISP should be able to fix any signal issue for free. If not find a new ISP because they are garbage.

there was no argument against else. my question was, when the entire if/else block takes up the full function block, should i just remove the else and just return in the 'if' part.

this SO post asks the same thing (shouldve searched there first):
http://stackoverflow.com/questions/9...f-if-else?lq=1



no one suggested this

Even motivated me to untangle some code today!

Brag: The web site I launched last week for the app I've been building is on the front page of builtwithbootstrap.com right now.

It's LPology, the one with the horrible looking screen shot. (note to self: text shadows look awful in a quick, untouched screen cap)

Also, you guys mind checking it out and telling me what you think? It's a little light on content at the moment, I didn't really expect any traffic at this point. But, I'd love some outside opinions since I basically develop in a vacuum.

Any and all opinions, criticism (especially), thoughts, etc. are very much appreciated.

Personally, I have never been a fan of the default bootstrap nav bar. I also think that the menu items in the nav bar look a bit odd right in the centre of the page (on my screen anyway).

The rest of the site looks good. Yes, content is low, but this is not always a bad thing, and the design is very clean.

Kyleb, from a programming standpoint, how do you like your MBP retina?

I like the bootstrap nav.

sdturner, I think the site looks clean and professional.

http://www.ericsink.com/entries/vcbe...tion_free.html

If anyone wants a free hard copy book on version control, this guys shipping them out. Got mine after about a week. Might make for some light bathroom reading xD

It's an overexpensive machine that runs a questionable Debian fork (given all the problems I'm having with it, anyway). I'm happy it was free. I would never spend the money they are asking for this machine.

This should be a longer post describing my reasoning. I'm currently about 70% done building this website. It's strange to write this: I don't get the appeal of web-development at all.

The concept, as far as I know, doesn't exist anywhere on the web, so it should be a pretty good business idea.

The stack:

Maven (Apache)
PostgreSQL
Clojure

So, with a decent idea and an interesting stack, I would have thought I would truly enjoy building this thing up, but no, I hate it.

It puts me in a weird mental place because if I take one step further, I can now understand web-dev companies that just use some drag-and-drop CMS templates and call that web-dev, which means that it's all about making money doing stuff you hate for a paycheck. I can't reasonably exist in a world like that, so I still fail to understand that mentality, however, it's almost like I can have empathy... nah, **** that; they suck.

*To clarify, this site is for me, not for someone else.

But the retina display is ****ing awesome. I will say that much. TN panels can't go away soon enough. Worth reading:

http://www.codinghorror.com/blog/201...evolution.html

External web dev is pretty infuriating; I hate front-end stuff. Roles that I've always flourished in are developing internal tools for data-related applications, because you generally don't have to worry about dumb stuff like tweaks to CSS. Most business units just want a decent access portal to analytics and/or data, and if you can solve 80% of their problems with simple work (which you generally can), they love you.

It's overcomplicated by the fact that most people want way more than is necessary. WordPress solves literally 95% of all demands on the web, yet everyone wants their own custom CMS or whatever bull**** nonsense they don't need. I worked at a lead generation company where we migrated away from PHP4 to Java (wtf) because they wanted an "enterprise-level" solution; they used Grails (lol) for the internal tools packages. It was a goddamned mess, and our biggest competitor was using WordPress and crushing us.

Good think it's in Clojure. If you ever decide to turn it into a business you'll attract plenty of good people imo

Yeah, the front-end is pretty minimal, and back-end is sort of MVC with a mind for extensibility. I would gladly use something like Drupal, but at least in my experience with it, once you make certain decisions, you're sort of married to to them, and this would be very bad assuming certain other needs would want to be met. I really want full control of the database and how it integrates with the code logic. If I was a CMS expert, I would probably think what just I wrote is plain silly, but for now, I don't like fighting through the abstraction barriers.

The front-end stuff is sort of irritating and sort of neat at the same time because the HTML on the page can only be generated through pre-processing, thus you wouldn't be able to say, create a front-end in notepad++ or Dreamweaver push it into Clojure and display the raw HTML on the screen.

Here's an example from my base:

The above is equivalent to include files in PHP. I could toss in a vector-map of arguments or pass in a function as arguments and have this one piece represent unlimited pages.

The following is not legal:

Since I am dealing with a simple front-end, there isn't too much to worry about in regards to integrating CSS. I just run it through SASS and that seems to take care of most of the BS for me, though I've yet to test it in IE7 and IE8. I really don't like bombast on websites, so the clean front-end isn't gong to require much CSS beyond the basics.

****

As for the code-base itself. I find that I have the most fun refactoring. It's
pretty amazing how quickly 200 lines can get wrapped up into like 40 lines, or simply changing a page layout or navigation interface can slice off 100 lines of code. Creating extra namespaces to separate the logic of the program parts is also a lot of fun. So, it's unfair to say that I don't enjoy the project, I just don't enjoy the website part of the project, if that makes any sense to you. I always think to myself: "How can I make sure this whole thing is only 200 lines of code?"

Yes, this is hopefully a good plus, but considering I'm the one creating it and I would never hire myself to create this site....

also
http://c2.com/cgi/wiki?GuardClause
http://c2.com/cgi/wiki?HandleErrorsInContext

Hunting down outliers in your dataset is such god damn fun!!!

Is it still that thing with the thing?

Nah, that was a different outlier.

I want to re-respond to this because it's a great post and I gave a non-answer the last time because I didn't want to go into religious reasoning. I only said that I rather get my own hands dirty and I feel that using a CMS (Drupal) would create too much difficulty because apparently I am a special snowflake. I can fully accept that the issue is my own inexperience and lack of patience. The way you describe the mess, I have to admit is sounds bad, and if I recall, you mentioned a while ago about working for a company in the Seattle area that is infamous for doing something...?

As far as WP goes, I have found myself suggesting people use it to build their own sites, but I always get the same answer: "WP is a blogging software, I'd rather use this program I'm using now: never mind that it's incompatible with any web host and has zero documentation or community support," which I explain to them to look at WP's main site and look at all the credit card and paypal plug-ins. I'm not saying it's easy to use, but if someone insists on building their own site, I would much rather suggest WP over Mount Drupal or Magento, with it being XML-based and all.

What I meant by my comment is this: I don't understand companies that can only use WP and then just use drag-and-drop UI to create everything. I think this smacks of un-professionalism, and yes, although I think doing this on any level is dishonest, at least, for the most part, they aren't charging more than $5,000 for a project. Sure, you won't see your site for about 6 months and it will probably suck, but at least they're somewhat honest in this regard.

A while back, I was researching a web-dev company here in Los Angeles, and I found it appalling that 85% of the companies I found were exactly what I just described, and many of them were charging something closer to $30,000.

I fail to understand why anyone would bother going to school for 4 to 6 years and then enter into these sorts of companies or specialize in CMS-based websites. Although I am dismissing a group of highly skilled developers with that comment, it's astounding that after all the interesting things you'd (hopefully) learn about computers and programming, that any sane person would want to sit down do stuff that superficially reflects programming. It would stand to reason then, that RoR, C#, and Django companies generally attract better developers because, although I am sure each company has a default CMS, each of those CMS's, as well as the front-end, is customizatable.

Have you ever heard of Agile Development in Word Press?

As far as a custom CMS, I don't know exactly what to make of that. I know some companies would prefer to switch around their site as they please, but I honestly think that allowing a customer to do this is really bad. Many of the sites I examined looked like Picasso drawings with all the different fonts, variable-styled bullet points, div sizes, and other muck they managed to put on the site. I am speaking not only from my research in web-dev houses, but also at the company I currently work at. I think that my, and everyone's, life at this company would be much easier it updating the site was three steps: Make the appropriate changes to a CSV, drag-and-drop the images to an FTP, upload the CSV. Boom: 300 updates complete in 1/2 day.

Even though PHP seems to be the bastard child of the web and reviled by many, I personally don't give a two craps what anyone uses for anything: just use a) what your comfortable with, b) the best tool for the job, and c) understand that a) and b) are non-inclusive.

As far as using Grails / JVM, I can understand a decision like this -- but not this one -- if they were actually deploying JVM-based products to desktop apps for enterprises. As I understand it, the whole point of these new-fangled JVM languages is to create enterprise-compatible programs that can be deployed with little resistance because so many companies use JVM for various reasons. But to rebuild a site like this does seem really stupid. But then again, so doesn't Oracle, SAP, Peach Tree, et.al. Why bother when you can just use LAMPP? The only difference is, that it sounds like your company just went into the new direction based on religion and not on business logic. Why would anyone rewrite their site in brand-new, untested, and unpopular language and framework?

I do agree that data-backed programming is more enticing and interesting: it's a good balance. I suppose though, that a good house will have a strong separation of duties. Unless someone wanted to come work on this site: I would just tell them to rewrite the whole thing and be glad they have a job. j/k.

This is a really good read. Actually shows the else / comment idea I mentioned, but also touches on pushing things out into a separate function, etc....

http://programmers.stackexchange.com...ad-programming

Don't want to clutter up the other thread but since Erlang was mentioned as a web development language (I agree FWIW) has anyone investigated the available options?

I haven't gotten much further than reading a couple of stackoverflow posts. Seems like the frameworks to chose from are Erlang Web (the notion of it being JAVA enterprisey is a prett big red flag for me) and Chicago Boss (which kind of seems like one dude doing his thing but seems better suited for me). EarlyWeb seems to be abandoned since the dude went to Facebook, Nitrogen is mostly frontendish stuff but I've read only good things about it.

Mochi Web isn't an entire framework but probably a good starting point.

Yaws as a webserver just to mention it.

I guess cliffnotes..I'd just pick Chicago Boss hope for the best and hack away :P

Just because you use a CMS as the basis for web development doesn't mean everything you're doing is mindless drag and drop. All it means is that you've taken away a lot of the repetitive drudge work of web development (eg, filtering for security, templating). The actual interesting parts of the development - creating the unique solutions for this client - are still there.

If all you're doing is creating yet another marketing website, then there's nothing interesting there, no matter what system you use. You want to get it done as quickly and cheaply as possible to better serve the client - and focus your time and effort on the things the matter to them (the design) and provide them with an easy-to-use editing interface.

I always thought the term CMS was pretty strange. By definition it's a content management system but when you use a CMS you're not just managing your content. You're managing and creating your entire site's functionality and output.

It seems like a mistake to me to try and use a CMS to mold your entire site because you'll spend a lot of time working within the limitations of the CMS and to do anything semi-custom you have to learn something that's only applicable to that specific CMS (little value in the grand scheme of things).

Also when you start really tinkering with stuff because "hey, my CMS is good, it lets me manually do whatever I want" then you lose the entire appeal of what a CMS does because it's no longer this generic thing that lets you mix and match plugins/modules anywhere. Now it's highly tuned to your specific site and is unusable anywhere else without modification.