Ask HN: Beginning freelance web developer: no clients - portfolio?

Since I'm constantly reading that a freelance web developer without online portfolio of websites/apps developed for previous clients is a big No-No, I was wondering how to get away with this chicken-and-egg problem of having no clients having no projects done for clients.

My plan so far is to deploy my websites/projects that I'm currently developing (and will develop) to my personal site and use them as a "fake" projects with the idea that they are actually "real-world". Let me explain a bit about what projects I'm planning to develop:

1. E-Commerce site

* Features: product catalog (departments, categories), product attributes and images, product reviews and recommendations, admin panel, shopping cart, customer/user accounts, ordering pipeline, credit card transactions (DataCash, Authorize.net), amazon.com web services

* Technologies: PHP (from scratch - in order to learn), MySQL (stored procedures), Smarty templates, AJAX (shopping cart updates), keyword-rich URLs (from scratch).

2. Community site for people with shared interests (let's say books)

* PHP + Zend framework (mainly to learn how to develop an app from scratch with Zend), Unit testing (PHPUnit), JavaScipt (jQuery), Web services (amazon, google calendar, youtube)

3. Another E-Commerce site

* This time developed with Zend Framework for the learning experience

4. Custom CMS

* Zend Framework

5. Music mashup app

* Zend Framework, Web services (RSS, amazon, youtube)

6. Blogging app

* Zend Framework, JavaScript (prototype, scrip.aculo.us), Smarty templates, Image gallery, Google Maps

My plan: From these, the #1 (the e-commerce site from scratch) is 2/3 done, I'm learning a ton... With the help of a good friend (and beginner web designer), we'll beautify the site, but we're wondering what kind of e-shop (regarding the types of products) is most suited for a portfolio in order to impress future clients?

The million-dollar questions:

* What do you guys think of my plan for getting complete applications and experience in my portfolio?

* Which of the above projects that I've planned would you scrap, or which would you prefer?

* What about Drupal: does it worth it to create a couple of full-featured (a restaurant site, image gallery, job board, etc.) (but still "fake") sites with it in order to show that I have these skills. Or generally, developing with Zend+JS as a sort of "specialization" will be more viable for my goals?

It is essential to NOTE that I plan to start bidding for projects on the freelance site (elance, odesk, etc.) and use my "fake" portfolio of home made projects in order to get work.

I also think that you are suffering from the classic symptom of trying to get everything just right before starting. When building a web app you want to get that thing up as fast as possible, even if it's ugly. Build a minimum viable product and connect with customers each step of the way. Test for interest, etc. You don't want to spend too much time getting everything correct just to find that you wasted your time because nobody cares for your idea.

Take a similar approach with web development. You could start out banging out sites in Drupal or Wordpress and each of these content management systems require more configuration than actual programming. You can learn these quickly and with minimal to no PHP. Your interest may be more in building web applications from scratch and there is a market for that, but you don't have to start out there. I know of plenty of people making a living strictly from building sites using already available content management systems.

Look through sites like Odesk and Elance for the types of jobs available. You might find the occasional gem in these places and even land a regular client. Don't feel bad for your lack of experience, you just need to be able to sell yourself and communicate well. Trust me, there are far worse people actually landing the jobs there.

Make sure to network through forums, Twitter and other locations where fellow developers and designers hang out. Twitter is especially good for getting a regular voice in the community and getting your name in front of people. Help out people in Twitter and forums and you will occasionally get someone hitting you up for doing some work.

As you are connecting with the community, look for ideas for building add-ons which might be in demand. This is a great way to give back and get your name out there. Offer some simple add-ons for free, but you might be able to make money by selling other add-ons.

As you beef up your freelance business and development chops, you can start moving more in the direction of building applications from scratch and perhaps working with other tools. There is a lot of exciting new things going on in the world of web development. I have been playing with Node.js a lot lately and it's a new community where you can more easily gain a good foothold as a pioneer as opposed to PHP. However, the market for this may not be as easy to get into at first.

There are also lots of methods I haven't mentioned or even thought about. Another big part of business is strategy and tactics. Figure out where you want to be and be creative in formulating a plan to get there. Do this and you might be able to leapfrog (greater income) people who have been doing this for a lot longer than you have.

3 months for learning web dev is short, then there's all the nonsense like configuring servers you're going to have to learn too (which will be a much bigger time sink than you would guess). Might as well get paid while you learn it. Hell I've done an access DB for someone and I'd never touched it before, relied on my SQL skills to get going. They were the ones insisting on Access btw, not my recommendation :)

It's good to know what you don't know, but don't let it scare you out of getting going.

The only other thing I can add is local networking. Start going out to local business events. I got work the first one I went to. Obviously won't happen like that all the time, but I've come across 3 or 4 opportunities in the 12 months I've been doing that and I'm crap at networking, only costs me 2 nights a month and it's usually fun. Good ROI in my book.

Oh, and be careful of underestimating how long something will take.

My question in regard was basically whether these "fake" personal projects can convince a prospective client that I'm for real. That is, if these personal projects are not on separate domains (like it will be if they were for real clients), wouldn't they look superficial, and yes, fake?