What’s uploaded, Doc?

In a nutshell, MD + PHR = ISP? In other words, when a medical doctor or clinic starts serving personal health records from their office, are they starting to turn into an Internet service provider?

Physicians in Ontario and British Columbia are blessed with access to the OSCAR system for EMR’s (That is, Electronic Medical Records.) From the IT perspective, there is an excellent (detailed and accurate!) OSCAR installation guide for setting up on the long-term-support edition of Ubuntu. It appears to me that OSCAR provides our physicians with good reliable free/open-source software so they can focus on looking after our health.

Whereas OSCAR provides convenience for the practitioner, MyOSCAR (related free/open-source software) gives access for the patient. Moreover, installation again seems straightforward, particularly in the case where a MyOSCAR server will be side-by-side with an OSCAR server. So physicians wishing to provide the best possible service are setting up MyOSCAR.

My concern is that when MyOSCAR is installed in a medical office, the ante goes up in terms of server demands. Is it wildly dystopian to imagine medical offices as moving towards becoming ISP’s, dealing with bandwith-burning hypochondriacs and crackers starting to see medical offices as prime targets?

My suggestion to an interested doctor or clinic would be to seriously consider installing OSCAR if they haven’t already, but to seek an Internet Service Provider sufficiently trustworthy to serve MyOSCAR from the ISP office. The rationale is that medical doctors should not risk getting distracted by ongoing concerns about network security.

It is a subtle issue. Even aside from technical convenience of having MyOSCAR side by side with OSCAR, there is a possibility that in some jurisdictions patient records on MyOSCAR may have better legal protection with the server in the medical office. So I cannot insist that the MyOSCAR-at-the-ISP approach is correct, I can only raise the question.

Send in the CAD clones

Cloners support inter-operability, good for the user. A commercial software vendor wanting to force upgrades might happily introduce incompatibilities with earlier versions, whereas a cloner will thrive on cross-version compatibility.

The success of Autodesk (evidently a $5.5 billion company) has spurred a bewildering variety of AutoCAD cloners, many of which are members of the IntelliCAD consortium.

AutoLISP, the classic AutoCAD programming language, preceded development of reactors (Autodesk jargon for the callback mechanism used to implement smart objects in drawings). AutoLISP got reactors when Autodesk acquired Vital LISP, rebranded as Visual LISP, but later Autodesk reverted to calling it all AutoLISP. Confusing, yes, so what is my point? Well, an AutoLISP-capable clone might only provide the classic reactor-less version; such an offering may be eminently scriptable, but I would not call it programmable.

CMS XPandy is the cheapest IntelliCAD offering, at a price of zero/adware. Yes, XPandy is a full CAD program, impressive, and very scriptable. Not enough for my purposes.

Bricscad is the most prominent low-cost programmable AutoCAD clone. The vendor, Bricsys, is slowly breaking away from the IntelliCAD consortium, frustrated by IntelliCAD’s slow development and Windows-centricity. Bricsys has done major code refactoring with a view to a Linux port. In the meantime, the popular choice for serious CAD work on Linux is VariCAD.

There are some noteworthy free software CAD projects, some of which are QCAD, Archimedes, and BRL-CAD. These do not appear to me to be moving very quickly. We can safely say that one factor is demand being cooled by recent heavy price competition among commercial AutoCAD cloners.

All the fuss about AutoCAD compatibility puts too much focus on systems that model surfaces (as AutoCAD does), rather than solids. Meanwhile, SolidWorks and Pro/Engineer are rather costly. Bizarrely, very little seems to have happened with BRL-CAD, which appears to be a wonderful basis for both free/open-source software and commercial products. Wake up, people.

Truth about SEO?

The Truth about Search Engine Optimization is a quite new (2009) book by Rebecca Lieb. To start with, she is quite readable and even entertaining:

Some spiders fetch entire pages; others are easily bored and look at only some of the content.

Doorway pages have the potential to be portals to a sort of search engine hell.

Any SEO who claims to have special tricks or secret methods to achieve high rankings is lying. There are no tricks to this trade, or secrets. There’s just solid knowledge and diligent hard work.

Secondly, the book is accurate, little if any misleading or wrong information.

Thirdly, the physical presentation is reasonably good. An inexpensive book on nice heavy paper, and I rather like the layout.

Fourthly, her approach complements my own articles on SEO Basics. She writes well about SEO technology from a marketing standpoint, whereas I’ve written about SEO marketing from a technical standpoint. It is an interesting contrast. Whether or not you find my own articles useful, I would appreciate feedback.

Being somewhat of a bibliophile, I bought the paperback and not the e-book. Because the paperback lacks an index, the e-book would have advantages.

As a footnote, on querying “Rebecca Lieb” to Google, the 30 entries are all about her, which is a good positive statement about the author’s SEO credibility.

Google Analytics complement

Today I was pleasantly surprised to find that my page on Website Analytics ranked in a search for Google Analytics substitute. Now can I please get ahead of the curve:

Hello, Google? Check out my page on my page on Spider Tracking. Since CrawlTrack is a natural Google Analytics complement, my suggestion is that Google should support CrawlTrack.

Stats for new WordPress site

Stats:

  • 100 human visitors from 14 countries, according to StatPress SEOlution (using IP2Nation). Countries: USA, Bahrain, Canada, UK, Germany, Luxembourg, Netherlands, Spain, Russian Federation, Thailand, Argentina, Brazil, Poland, China. The true numbers are lower, as some of these visitors are actually bots, but the data is interesting.
  • 24 distinct bots recognized by StatPress SEOlution. The true number is higher, perhaps 30, with some bots being unrecognized.
  • 80 feeds.
  • 50 spam comments blocked by WP-SpamFree.

That is after being up only 37 days, and still low profile. Conclusions:

  1. Hurrah for WP-SpamFree!
  2. Compared with Google Analytics, StatPress SEOlution is great for real-time feedback and bot data. It does not completely replace GA, which provides additional help on searches and (apparently) more accurate information about countries.
  3. The feeds are evidently important, but (being new to WP) I know little about them as yet.

Of course StatPress SEOlution and Google Analytics have provided a lot of other interesting stats, omitted above because the stats are more site-specific and/or more complex to interpret.

WordPress SEO gotcha – Digg

Having content hijacked into framing (by a site such as Digg) means losing both traffic and PageRank. WordPress webmasters are

  • Very concerned about this type of problem, seen by huge popularity of SEO plugins; but
  • Fairly oblivious to this particular problem, seen by relatively low popularity of plugins that address it.

Current best practice is using frame breaker code, by a specialized plugin or else within HeadSpace (enable Site Modules -> Frame Breaker). However, specialized frame-breaking plugins have few downloads, and HeadSpace has far fewer downloads than other SEO plugins without this functionality.

On a side note, WP Super Cache makes a confusing suggestion to tell the world your server is Digg proof! WP Super Cache is excellent for bolstering a site for heavy traffic, but is no defence against Digg-style hijacking.

WordPress gotcha – revisions

Most WP users need a Revision Management plugin to work around a flaw in the WP user interface.

WordPress stores unlimited undo information in the database. That could be handy for a blog needing an audit trail, but otherwise it is a pain that the WP user interface gives no way to make a revision permanent in the sense of cleaning out the old junk. It is not so easy to know how much the old undo data is bloating the database or slowing response times.

The problem is acknowledged in the WP Codex page on Revision Management, where we see that deleting post/page revisions is left to plugins.

There are more than half a dozen different plugins dealing with this situation in various ways. A real annoyance, but somehow never a really hot topic, judging by the modest numbers of downloads. (The most popular has under 20,000 downloads.)

What should an executive know about SEO?

My series on SEO Basics is now a completed first draft answering that question and more, delving into

  • Analytics tools,
  • Getting the right things on search result pages,
  • Keywords,
  • On-site SEO,
  • What makes a website useful,
  • Web standards, and
  • The major search providers.

The idea is to provide an informative overview, steering a middle ground between big-picture marketing and tiny details, and focussing on the fun parts.

Search snippet bonsai

SERP (search engine result page) entries are getting more like miniature web pages. To the snippet bonsai master:

  • The main text snippet is the trunk,
  • Bonus links are branches,
  • Bonus images are leaves, and
  • Other bells & whistles are flowers.

Four pages about this are now a completed first draft, part of my Search Basics sequence. It is designed to educate my SEO clients/prospects, but may also be useful for other search engine optimizers and webmasters.

SEO jargon page improving

My original series of SEO management articles continues:

  1. Jargon page (SEO glossary) is better structured, starting to get some appropriate cross-references;
  2. Table of contents put in;
  3. Intro to search engines is more concise, with a good analogy;
  4. Concise description of how Google struck gold;
  5. Various other clarifications.

The url http://johnmacphail.com/seo-basics/ should be stable, if anyone cares to link to it…