Eleven years – has it really been that long since we started TidBITS? Adam looks back at what’s changed and what’s remained the same in the Macintosh industry. Dr. Ron Risley concludes his article on communicating with your doctor via email with thoughts about minimizing the risks of using electronic communications. We also cover the important releases of Mac OS X 10.0.1 and the public beta of Retrospect Client for Mac OS X.
Brief Network Outage 17-Apr-01
Brief Network Outage 17-Apr-01 — We normally wouldn’t stress about a short downtime, but Qwest tells us they plan to take my 56K frame relay connection down for a software upgrade on Tuesday 17-Apr-01 from 4 A.M. to 7 A.M. (U.S. Pacific time), which coincides with one of our busiest times of the week. All services at tidbits.com will be unavailable during that time. [ACE]
Mac OS X 10.0.1 Available
Mac OS X 10.0.1 Available — A mere three weeks after the initial release of Mac OS X 10.0, Apple has released 10.0.1, a 4.1 MB update available via Mac OS X’s Software Update control panel (which itself requires a 700K update before it gets the 10.0.1 Update). According to Apple, improvements include better support for third party USB devices, improved performance, better overall application and Classic-mode stability, enhanced support for iTunes, and support for the Secure Shell service (albeit through an older version than is commonly available elsewhere). [ACE]
Retrospect Client Public Beta for Mac OS X
Retrospect Client Public Beta for Mac OS X — Dantz Development has released the public beta of Retrospect Client for Mac OS X, making it possible to back up Mac OS X machines via a plug-in added to Retrospect 4.3 on a backup server running Mac OS 8.x or Mac OS 9.x. Since Unix permissions of files are lost for any files backed up while a Mac OS X-based Mac is booted into Mac OS 9.1, this public beta, despite its raw state, is an important step in making Mac OS X usable on a production machine. Limitations include no support for Retrospect selectors, disabled notification and backup server preferences, no clock synchronization, no countdown alert before backup, incomplete support for UFS volumes, no uninstaller, and an expiration date of 01-Jul-01. Complete restores also require booting from another volume that has the Retrospect Client for Mac OS X installed. A number of problems have already arisen on Dantz’s support mailing list (the only support option available right now) – I strongly encourage anyone reliant on Retrospect to subscribe by sending email to <[email protected]>. The 444K download requires your Retrospect license code or registration number, and you’ll need an available client license to log in to Retrospect Client for Mac OS X. [ACE]
What’s Up, eDoc? Emailing Your Doctor, Part 2
In a previous article, I presented some of the reasons why doctors and patients would both benefit from more widespread use of email, along with some of the problems inherent in doing so. This week I will cover some steps that doctors and patients can take to ensure safety and minimize the risk of miscommunication when corresponding via email.
What’s Being Done? Some brief guidelines from the American Medical Association and a more detailed analysis by the Massachusetts Health Data Consortium have begun to address some of the issues inherent in doctor-patient email communication. By understanding and following these guidelines, physicians and patients can use email effectively while appreciating some of its limitations.
<http://www.mahealthdata.org/mhdc/mhdc2.nsf/ e214ac63ff65c87e852564580073a9fd/ 4a7c6d398962159785256759006a1113? OpenDocument>
In addition, Federal regulations known as HIPAA (the Health Information Portability and Accessibility Act) will likely put significant constraints on medical use of email in the near future. HIPAA will probably require patients to sign a written agreement to waive confidentiality before a physician can communicate any part of a patient’s record by electronic means without using strong encryption. Although the new attention to privacy is welcome, the regulations could also become a barrier to using email to improve patient-doctor and doctor-doctor communications, while making it far easier for institutions like insurers, government agencies, and employers to access and share the same data.
Personal Encryption — One solution to email privacy issues is personal encryption. PGP (Pretty Good Privacy) is a venerable suite of public-key programs which can secure email communications channels. It is available free of charge for personal use on Mac OS 8 and 9, various flavors of Windows, and Unix/Linux platforms; commercial versions are offered by Network Associates. Freeware open-source versions are also available. PGP distributions are limited to the United States and Canada because of restrictions on the export of strong cryptography. There is an international version, PGPi, distributed outside the U.S., as well as a fully compatible Gnu public licensed counterpart, Gnu Privacy Guard (GPG or GnuPG) that has been ported to Mac OS X.
The problem with PGP is that it requires both the sender and recipient to install and comprehend PGP or compatible software, generate keys, and reliably distribute their public keys. Key management and distribution can be a pain. Though recent versions of PGP have come a long way in improving their usability (and are well integrated with a number of modern email clients), PGP is still a long way from being user friendly. Two of my patients have actually gone to the trouble to install it and use it, but they are a distinct minority.
The Guerilla Factor — While institutions drag their administrative feet getting patient care systems running, and government agencies struggle to create byzantine regulations, the Internet, as usual, surges ahead. Instead of fostering instant communication of private personal information between insurers and institutions, the Internet just might make it possible for patients and their physicians to take back ownership of their personal information.
Regular TidBITS readers might recall I was bitten by the server bug a couple of years ago, and set up my own corner of the Internet using a broken PowerBook 5300cs. One of the advantages I saw at the time was that, by hosting my own server, I could at least ensure that email sent to me by patients didn’t sit on a commercial server until I picked it up.
I have since recycled a number of old Macs, and my junkyard server farm has grown to six machines. Although most run the Mac OS, I decided that security and privacy issues with medical communication could only effectively be handled using Secure Sockets Layer (SSL), the technology behind secure Web commerce sites. I am on an impossibly tight budget, and there are no free or low-cost SSL servers for the Mac OS (although that is changing with the release of Mac OS X, it requires an expensive machine). So I loaded LinuxPPC on an old Power Mac 7200 and installed Apache-SSL (a process that, despite my experience as a Unix system administrator in the 1980s, revived my respect for the ease-of-use of the Mac OS). I had to fork over $125 to Thawte for a secure server certificate (more than the cost of a complete 7200 from TidBITS sponsor Small Dog Electronics!), but the Guerilla Physician Project now has a secure server.
What is the Guerilla Physician Project doing? Confidentiality is extremely important in the treatment of Huntington’s Disease, an inherited genetic disorder, as it is possible to test as genetically positive yet have no symptoms. Someone who might not develop any problems for decades will nonetheless find themselves unemployable and uninsurable, yet testing can be valuable both in helping patients plan their futures and in preparing for early intervention when problems develop. I have set up an electronic communications network for the Huntington’s Disease treatment team here. Since the team is a multidisciplinary group involving state and county agencies as well as the university, it would have taken years to get all their IS people together to design and approve a system for secure communication.
Instead, I was able to bring the system online in a few weeks using the Guerilla Physician server, along with open source bulletin board and chat software, and a lot of sweat equity – but with no budget requirements whatsoever. Users of the system need nothing more than an SSL-equipped Web browser, which means they don’t have to load any special software onto machines whose program suites are often tightly controlled. Unlike commercial or corporate systems, the data is never in the hands of anyone who is not a licensed health care provider on the treatment team.
The Guerilla Physician is expanding, with new projects to help integrate mental health care in the diverse reaches of rural California. I am also coding a Web-based email system that will enable patients and physicians to communicate using PGP encryption without going through the difficult and sometimes tricky process of installing and using PGP on their own computers. Once this is in place, the potential for a truly private and secure distributed electronic medical record – shared only between patients and their physicians, will be a step closer to reality.
The Future — The recent release of Mac OS X and the proliferation of broadband Net access might well lower the threshold enough that more doctors will be able to host services like the Guerilla Physician. Medicine is an odd pursuit, in that it can combine the most intimate of personal interactions with some of the world’s largest and most impersonal institutions. My hope is that the distributed power of the Internet will be used to restore privacy instead of compromising it.
[Ron Risley is a family doctor, psychiatrist, former communications engineer, and inveterate hacker plying his trades in Sacramento, California.]
TidBITS Goes to Eleven
Today marks the beginning of our 11th consecutive year of publication, finally giving us the right to play off the famous Spinal Tap quote, "These go to eleven." Previous anniversary articles covered our basic history, motivations ("TidBITS Nets Ninth Anniversary"), and the lessons we’ve learned over the last ten years ("Lessons from Ten Years of TidBITS").
To crank up the volume this year, I decided to read through the TidBITS issues from ten years ago, when we’d had a chance to smooth out the rough edges of our first year of publication. What I was most curious about is how things have both changed and stayed the same over the last ten years, and as I read, the names of people, products, and companies came flooding back. Here then are some of the high points of that year for both TidBITS and the Macintosh industry, with some thoughts about how these changes have rippled forward to affect today’s world.
If you’re interested in browsing through history like this (and I’d challenge other publications to make their entire publishing history available online), the easiest method is to download all of our setext files from ftp.tidbits.com and use Easy View to page through (Easy View hasn’t been updated in years, but it’s still functional and fast on my Power Mac G4 running Mac OS 9.1). If nothing else, think back to where you were in April of 1991 as you read on.
Changes in TidBITS — The most striking change I noticed was how short our articles and issues were and how similar they seemed to the kinds of discussions that now take place in TidBITS Talk. In particular, the MailBITS section we now use for short news items really was devoted to reader mail then. Reviews were also shorter, though we tended to publish them on their own rather than in our regular weekly issues. Eventually I realized why we’d lengthened our articles.
When Tonya and I started TidBITS, we intended articles to be short news summaries. I proved unable to resist commenting, but back in 1991 and 1992, I didn’t know all that much about the Macintosh and the industry. I was willing to try anything, and I experimented far more than today, but I simply lacked the background data with which to fill in articles. Additional evidence of my ignorance comes from the frequency with which I made mistakes and corrected them in the next issue. Now, since I’ve been at this for eleven years, the information I can call to mind on almost any topic has increased, making it easier to add useful details. Mistakes still happen, of course, but our editing and fact-checking skills have improved tremendously.
The scarcity of information also played a role. This was before the rise of the Web, and I gleaned information from personal email, mailing lists like the Info-Mac Digest, Usenet news, discussions on AOL, trade publications like MacWEEK and InfoWorld, and spec sheets picked up at Macworld Expos. With generally incomplete information, it was hard to write long pieces, and mistakes were easier to make. Rumors played a larger role, since solid information was scarce and the online world wasn’t large enough to merit much concern from Apple. Everything was smaller and simpler, with fewer models of the Mac, less software, fewer companies, and many millions fewer Macintosh users. I was especially struck by how we’d occasionally direct a comment at a specific reader, and once we passed on a note asking for information about three Macs stolen from a Maryland warehouse. I hesitate to think how many Macs have been stolen in the last ten years.
Another factor was our publishing medium – a HyperCard stack that merged itself with an archive of previous issues each week. Reading in HyperCard was a bit clumsy, so we switched to setext (structure enhanced text) format in TidBITS-100, at which point there’s a definite uptick in issue sizes.
Personal & Personnel Changes — We’ve always tried to run TidBITS on a personal basis, since that’s how we interact with the world, and personal comments in TidBITS issues from 1991 and 1992 brought back many memories. Those were pivotal years for us: Tonya and I married in June of 1991; we moved from Ithaca, New York, to Seattle in August of 1991; I went to my first Macworld San Francisco in January of 1992. From the way I wrote about the move and Macworld, I was much younger and geekier. I was also scared, although I don’t think it showed in the issues, and I was desperately trying to prove myself in many ways.
All those changes added up to a stressful time. We were young and resilient, but finances were tight. My primary lifeline turned out to be the Internet (via a 2400 bps modem connection to a UUCP system run by a guy who later helped start Northwest Nexus, our current ISP). TidBITS gave me something to concentrate on at a time when I needed focus in my life. That was when I created our sponsorship program so TidBITS could start to earn its way. When I announced our first sponsors in July of 1992, TidBITS was among the very first to carry advertising of any sort on the Internet. If we’d patented the concept then, perhaps we too could be going bankrupt today.
The sponsorship program took years to reach today’s level, but sticking with TidBITS in 1991 and 1992 gave me enough of a reputation and writing confidence to write the first edition of Internet Starter Kit for Macintosh in 1993. It evolved into a series of best-selling books, changed the stresses in our lives significantly, helped hundreds of thousands of people get on the Internet, and, not inconsequentially, swelled the ranks of TidBITS readers.
Along with the navel-gazing that comes when reading one’s own past writing, I was struck by how many people we still know were involved with TidBITS in those days. Contributing Editor Mark Anbinder has helped out since the very beginning, and he shepherded TidBITS through the time I spent without decent Internet access in Seattle (we may complain about slow DNS changes now, but back in 1991, it took a month for the UUCP maps to update). Contributing Editor Matt Neuburg first wrote for us in TidBITS-095, providing a full issue review of Eastgate Systems’ hypertext editor Storyspace. And Glenn Fleishman, with whom we later started NetBITS, first appeared back in 1991 and 1992 as well. A number of other names from those years – Marshall Clow, Edward Reid, Mark Nagata, Paul Durrant, Larry Rosenstein – frequently appear today in TidBITS Talk.
TidBITS Services — These days, we provide numerous Internet services, from our basic Web site to our searchable article database, and we’ve evolved a system that distributes these tasks among eight different Macs. In 1991 though, we weren’t even distributing TidBITS via a mailing list, just via Usenet news and the Info-Mac Archive.
Then, on the issue that coincided with my 24th birthday, I announced an email-based file server run via ICE Engineering’s UUCP program uAccess. Basically, you could send a specially formatted email message to a fileserver address and it would return the file to you via email.
I had only a 2400 bps modem then, so I couldn’t serve issues or run a mailing list via uAccess. However, a short while later, Alvin Khoo at Simon Fraser University offered to host a mailing list for TidBITS. The interest in subscriptions almost instantly overwhelmed Alvin’s homegrown mailing list server, and within a month Mark Williamson at Rice University offered to host our list on Rice’s LISTSERV. A few weeks later, the system administrators at Simon Fraser decided they didn’t want to baby-sit Alvin’s list, so we moved everyone to the Rice LISTSERV, where we stayed until we set up our own server running ListSTAR in mid-1996. Interestingly, nineteen people who subscribed to TidBITS in the brief time it was at Simon Fraser still receive TidBITS via those subscriptions today.
Looking Forward from 1991/92 — Although ten years is a long time, many of TidBITS’s old themes echo today. Even back in 1991 we emphasized backups, with reviews of Retrospect and DiskFit Pro. Other recurring topics include the use of multiple monitors, coverage of digital cameras, HyperCard (which the International HyperCard Users Group is trying to convince Apple to carbonize for Mac OS X), Mac OS usage tips, and advice for donating old computers
I was also pleased to see that TidBITS paid attention to international issues related to Apple and the Macintosh. Our resources were limited and we didn’t have volunteer teams of translators, but a number of readers contributed news from other countries and helped with distribution around the world.
Some topics were grounded more in desire than reality. Wireless networking appeared a number of times, since Apple was agitating with the FCC for some wireless spectrum for Data-PCS, and a number of other bits of news came along shortly thereafter. It’s a little sad that it took until Apple’s release of the AirPort products in July of 1999 for wireless networking to become real for most people.
A painful read from 1992 was the article in which I laid out the basics of preemptive multitasking, protected memory, multi-threading, and dynamic link libraries, speculating that Apple would be building these features into the Mac OS at the same time as the move to PowerPC. Shared libraries and multi-threading came to the Mac OS some time ago, but it took a full eight years and numerous dead-ends for Apple to bring preemptive multitasking and protected memory to market with Mac OS X. It didn’t need to take this long – all the stillborn options failed due to management and leadership failures, not overwhelming technical difficulties, and the main reason Mac OS X has seen the light of day is Steve Jobs’s management aggressiveness. Since the move to the PowerPC chip, no other Apple CEO had the guts to force developers to rewrite or even recompile their applications for a new Mac OS.
The biggest story in 1991 and 1992 for TidBITS was the 24-bit ROM debacle that started in May of 1991. The ROM chips in the SE/30, IIx, and IIcx limited those machines to 16 MB of RAM even though Apple advertised them as being able to access 128 MB (RAM cost about $40 per megabyte then, compared to 35 cents per megabyte now). Jim Gaynor (then at Ohio State) started a mailing list to discuss the problem, but politics there forced him to shut it down, and I ended up coordinating an open letter to Apple asking for a statement about the ROM problem. I gathered 576 signatures, sent the letter off, and was thoroughly ignored by Apple management. (See? Some things never change.) By June, Connectix solved the problem with MODE32. At first MODE32 cost $170, but by September the pressure on Apple – some of it legal threats – resulted in Apple’s licensing MODE32 from Connectix, distributing it for free, providing official support, and even reimbursing those who had paid for it. It was a major fiasco, and lest we think such a thing couldn’t happen again, think of the recent firmware update brouhaha, where an independent developer stepped up to solve RAM problems that Apple should have addressed.
Prescience in Action? Some realities of today’s computing environment appeared in TidBITS in fictional form. For instance, I fabricated an article about a distributed computing product (complete with a suspiciously familiar supercomputer ad slogan) in our April Fools issue in 1991 and followed up with discussions in two later issues about how it wasn’t so fictional. Today, we have the [email protected] project (and many others) pulling together vast computing resources from all across the world.
The 1991 April Fools issue was good (it also predicted IBM buying Lotus, which happened four years later), but I’m equally as fond of our 1992 prank. I wrote about modem-based remote backup using Retrospect, something that has become commonplace over the Internet with recent versions of Retrospect and the BackJack service. I also talked about how Microsoft would be porting its applications to the NeXT operating system, something that has come true after a fashion with the carbonized version of Internet Explorer 5.1 for Mac OS X. And in an article about an upcoming third-party Finder replacement, I suggested it would enable alias creation by holding down a modifier key while dragging (got that one!), that the Standard File Dialog would boast an outline view (much like what later appeared in Apple’s Navigation Services), and that it would have "super folders" that sound a bit like Mac OS X’s packages (collections of files that appear to the user as a single file).
Quote the Raven, "Nevermore" — Although many technologies, products, and companies have evolved through to modern times, others have fallen by the wayside. Most telling was the issue from Macworld Boston in August of 1991, since almost every product mentioned died long ago. Remember Lotus Jazz, Claris Resolve, More After Dark, Abaton InterShare, Spectre, Hand-Off II, Outbound Macintosh laptops, and the NewTek Video Toaster?
We wrote frequently about compression software, since 1991 fell during the days of the compression wars. Hard disk space was expensive, and compression helped conserve what little space you had. Archiving programs like Aladdin’s StuffIt Deluxe, Bill Goodman’s Compact Pro, Salient’s DiskDoubler, and Alysis’s SuperDisk were the first wave, followed shortly by transparent compression programs that worked at a deeper level, such as Salient’s AutoDoubler, Aladdin’s SpaceSaver, Alysis’s More Disk Space, and Golden Triangle’s driver-level DiskSpace. The speed and extent of innovation in this space was a testament to the power of competition. But as hard disk prices dropped and capacities rose (we wrote about $200 88 MB SyQuest and 90 MB Bernoulli cartridges then; now an 80 GB IDE hard disk costs roughly the same amount), the speed hit and compatibility issues raised by the transparent compression programs eventually drove them to extinction. Of all the companies and products, only Aladdin and their StuffIt Deluxe product remain active.
Ten years ago, viruses were a much larger problem on the Mac than they are today, and I’m happy we haven’t needed to write about new viruses as PC publications have. That’s due mostly to the lack of significant new viruses in the Macintosh world (pesky macro viruses remain the main trouble) and also because the anti-virus programs can now update themselves automatically, whereas in 1991, every new virus required a revision of the anti-virus software, such as John Norstad’s venerable Disinfectant.
Finally, we frequently wrote about pricing issues back in 1991, passing on news of deals, special offers, and unusual bundles. It made sense then, since there few other resources that could disseminate news before a special deal expired. We gradually phased out such news, since it felt like we were just providing free advertising for vendors, and there were an ever-increasing number of deals. In October of 1995, we started a sister publication called DealBITS to provide nothing but special offers, but it was several years ahead of its time and didn’t play to our strengths – a much better job is done by our friends at dealmac now.
Moving On — Although paying attention to history and learning from it can be useful, that’s different from living in the past. When you read exactly what was happening in the Macintosh industry ten years ago, you see the goal then, as now, was to move the Macintosh platform forward. There may be hiccups, missteps, and even serious dead-ends along the way, but there’s always a basic drive to improve, enhance, and try new approaches to computing. Individuals may choose (quite reasonably) to hop off the industry bandwagon or to change to a different path, but at this moment the Macintosh industry is larger and arguably more vibrant than at any point since we started TidBITS in 1990. My fervent hope is that today’s Macintosh community, like the community of a decade ago, will advance the platform in an active, constructive way, and continue to make the Macintosh the first – and best – platform for personal computing into the future.