Ask Soylent: Help Me Build Out the Global Computer Employer Index?

(This Soylent News article that I submitted today is a good start at my crowdfunding pitch.

It is commonly claimed that there presently is a shortage of one-half million software engineers; that alleged "shortage" is expected to increase to one million by 2020. If that's really true, why were so very many coders, web designers and 3-D wireframe artists dining with me at the Portland Rescue Mission and Blanchet House of Hospitality?

Consider the Stock Market:

I am quite certain that I know how to "Time The Market" for the stocks of certain specific kinds of companies. A friend asserts that cannot possibly be the case, as "every stock market investor has the same complete information." While true in principal, few if any investors avail themselves of truly "complete" information. They simply do not have the time. I've stumbled upon a Leading Economic Indicator that is available free of charge to everyone but known to very few investors other than myself. This Leading Indicator is widely known - but to other kinds of people than investors.

The perceived - and only perceived - shortage of coders despite our widespread unemployment is explained by a similar ready availability of information that we simply do not have the time to find.

Agency Recruiters typically require a commission of $30,000.00 to place just one mid-career Software Engineer. Job board posts are expensive. Candidate contact info from resume banks is exorbitantly expensive. Most who seek work don't really know how to find it. Most who want new staff don't know how to get it. Job seekers commonly submit hundreds of applications before getting an offer. Hiring managers commonly receive one thousand resumes for each open position.

Yes the complete information is "available" to all, but few other than myself have the time to turn it all up. As a consultant I am always looking for new work, so I know lots of clever, unobvious - and quick - ways to find new clients. (I could tell you how but then I'd have to kill you.)


I plan to launch a crowdfunding campaign soon in hopes of raising enough money that I can devote myself full-time to the Index for an entire year, as well as to hire two entry-level web designers and to cover office rental, equipment and the like.

Getting funded obviously depends on how effective the Index already is at the time of my crowdunding campaign's launch.

I seek your help in finding employers who are likely to hire computer professionals - not just Software Engineers, but also Electrical Engineers, Quality Assurance, Technical Sales and Marketing, Tech Support as well as Product, Project and Personnel Management.

(More below the fold.)

Intel and Apple employ both software and hardware staff; Moon Valley Software hires only for software. The Bonneville Power Administration employs mostly hardware people.

The Global Computer Employer Index is not a job board. Job board posts go away when the positions are filled. Most open positions - in every occupation - are never advertised. My Index links directly to the "Jobs" or "Careers" pages at each company's own website:

Adding your company's listing is completely free of charge and always will be!

Two web designers are working towards mobile device responsiveness. You could help with purely conceptual suggestions as to how such a responsive site could work. I'll discuss your ideas with my designers then get them to do the heavy lifting.

I have a bunch of LibreOffice spreadsheets with the links for roughly a thousand companies in a couple dozen countries. If you're looking for a job in a specific city, please tell me where and I will publish what I already have for that city as well as focus my efforts on finding new listings there.

My coverage even for such Tech Meccas as Hillsboro and Berlin is woefully incomplete. You can help by giving me the names of any software and hardware employers anywhere.

While the complete Name - Jobs - Contact - City info is quite helpful, all I really require are the names of the companies.

Is your employer listed?

I get the most mileage out of mid-size companies such as Magenic that have a modest number of different locations, as well as companies in such places as Couer d'Alene, Idaho that are not known as technology hubs:

My most-popular page is that for rural and sparsely-populated San Luis Obispo County, California. I found it damn near impossible just to locate potential employers when I lived there. The local message boards turned up naught but complaints about the great difficulty of finding coding work.

I will find the multinationals and Fortune 500s on my own. Such companies may have hundreds, even thousands of locations. I hope to get machine-readible listings of their locations directly from their Human Resources departments.

I do not yet list Silicon Valley in part because everyone there knows how to get a job, and in part because I shall give priority to the smaller companies that do not have the wherewithal to pay for headhunters or job board posts.

You could facilitate that by sending me lists of attendees or exhibitors from conferences and trade shows. If it's more convenient to send me printed listings, snail-mail them to:

Michael Crawford
650 NW Irving St
Portland OR 97209

I will pay you back for the postage as well as your effort when then crowdfunding comes through.

I attended the '89 USENIX Conference in San Francisco. So did Every Headhunter Who Walked The Earth. The day finally came that I grew weary of getting so very many phone calls at work so I asked Judy Golden to set me up. She sent me to Working Software of Santa Cruz, the best job I ever had. Company President Dave Johnson asked Judy how she "found such great local candidates". Her reply?

"It's a trade secret."

I won't monetize the site directly. I aim only to encourage "organic links". Those are inbound links that are placed voluntarily - and free of charge - by live humans. Such links bring me more-frequent inquiries from potential clients. Were I to get any coding work at all I could easily pay someone else to work on the site:

Each page footer links to my own consulting firm. Measures of link popularity such as Google's PageRank are - for want of a better word - "transitive" in that Link Popularity "flows through" intermediate pages to more-distantly linked pages. That is, were Rob Malda to link to Rusty Foster, and were Rusty Foster to link to me, then Rob's link to Rusty would result in some of Rob's PageRank increasing my own.

You need not point out the site's many petty problems: I and my designers will fix them Real Soon Now.

It's not yet automated. The most time-consuming work is finding the companies, then - quite stupidly - struggling desperately to somehow dig up their Careers or Jobs pages. Many open positions are advertised on the job boards as well as the companies' own websites - but those "corporate job boards" are not linked from anywhere!

Thus my first morsel of automation is a brute-force search through the common locations:


I will not list staffing firms or contract personnel agencies such as Kelly IT Services, Manpower Professional or Oxford Global Resources. I'm not really hoping to drive them all into insolvency - they are welcome to use my Index as a source of leads. The quality competition the Index provides will encourage them to return to the truly-useful "Value Added" services they offered when I was but a Clueless Newbie. Back in the day, Cupertino's then-tiny Oxford and Associates got me my first Mac coding job at Apple. Similarly-tiny XXCAL doubled my pay over a single weekend!

While I hold the copyrights to each page as a whole I do not assert ownership over the listings. (My understanding is databases may not be copyrighted.)

Slurp my entire site:

Once I fix all the nits, each page will be Valid XHTML 1.0 Strict. Each city's lists will be in a <table> element with an id="jobs" attribute:

<table id="jobs">
... payload goes here...

Find my Sitemap starting from my robots.txt. That yields my Sitemap Index. Parsing my Sitemap Index' XML yields the individual Sitemaps - presently just one plain-text list of URLs. Do not assume that list of URLs will always be there! Always start with the robots.txt:

I will soon replace the URL list with a proper XML Sitemap; then later an additional one - both listed in the Sitemap Index - with new and so frequently-changing pages in one Sitemap and the completely built-out - unlikely to change - pages in the second.

I will implement my RSS feeds with a cron job that slurps my site that way, as well as to emit raw XML for each city, county or parish, state or province, country and occupation. (I will ultimately cover many kinds of work other than technology.)

I Am Eternally In Your Debt.