» Archive for the 'Networks and Operating Systems' Category

Everybody Loves a Cloud

Wednesday, October 21st, 2009 by Jonathan Spira

Is cloud computing just the latest fad?

The IT industry loves a new buzzword, even if it’s just a new word and not a new concept. Cloud computing is the buzzword du jour and software vendors small and large want everyone to enjoy life in the cloud.

But is cloud computing really that new? Or is it a move back to centralized systems, reminiscent of the days when the mainframe was king with a dash of Software-as-a-Service added?

Some cloud resources are available as a utility (metered service, conceptually similar to a public utility such as the electric company) or as a service (billed monthly or annually regardless of usage). The concept of utility computing dates back to the 1960s and was in use in the 1990s for asynchronous transfer mode (ATM) networks. As for the cloud itself, as Amy Wohl recently reminded me, a quick look at network slides from the 1980s show clouds sitting above every network. I recall the term “telecom cloud” being used to describe VPN networks deployed by telephone companies because the routing for data traffic was, for all intents and purposes, somewhat cloudy.  And Larry Ellison of Oracle asked,  at a recent Churchill Club presentation, if it were true that companies such as Google were based on “water vapor.”

Needless to say, there are some clear and unequivocal advantages to cloud computing. A company can deploy IT resources on demand with little or no up-front cost. Therein lies the rub. While the barrier to entry (for end-users) is minimal, the usage or monthly fees continue ad infinitum, unlike licensed software for which there is only a one-time cost. Granted, larger organizations pay maintenance and support charges for their enterprise software, so this may be a greater expense to smaller organizations that have not paid such fees in the past. Companies that migrate a good part of their IT infrastructure to the cloud can eliminate significant costs for hardware and upkeep, not to mention IT personnel.

Cloud computing has also enabled new business models. Indeed, the rise of mobile computing and smaller smart devices (such as netbooks and smartphones) – devices with little storage and relatively few installed applications – may serve to drive the next generation of cloud computing applications. Cloud computing has also allowed software companies to develop applications tailored to individual consumers (as opposed to large companies) and this is yet another area where we are just scratching the surface in terms of innovation. It’s important to keep in mind that some knowledge workers within larger organizations may use applications from the cloud intended more for consumers and may ask for enterprise services from the cloud based on their experiences; others may sneak in applications (under the radar of their IT departments) that they need to do their work, making the distinction between enterprise and consumer cloud apps a bit blurry.)

Despite the hype, cloud computing comes with some baggage. There is no agreement on standards or one single architecture. In fact the one thing for which there is consensus is that there is no true consensus as to what cloud computing really is.

This doesn’t mean that managers shouldn’t investigate whether cloud computing might work for their organizations. But before they do, there are three factors to consider.

First and foremost is vendor lock-in, a concept familiar to many enterprise software buyers. Competing cloud providers have their own standards and formats, many of which are incompatible with one another. Indeed, even the simple (in concept) task of moving data from one cloud to another is fraught with peril.

Second is storage. While large vendors such as Salesforce.com have built up considerable trust with the user community, there are new entrants to the cloud arena every day that haven’t earned their stripes. Is your organization’s data secure? Are proper security precautions in place, both in terms of on line access as well as physical egress? For free services, what do the vendor’s terms of service allow it to do with personal data?

Last, and possibly most important, is the data stored in the cloud safe from loss? While cloud providers have gone down from time to time (Google’s Gmail service has had seven outages so far this year), the point of safety was driven home quite strongly last week by the T-Mobile Sidekick debacle, in which tens of thousands of users lost myriad personal data (including address books and photo albums, among other assets), while said data was entrusted to the aptly-named Danger subsidiary of Microsoft.

Jonathan B. Spira is CEO and Chief Analyst at Basex.

Google Apps Twitter Hack Raises Red Flags on Password Security

Thursday, July 16th, 2009 by David Goldes

One might presume that technology companies do a better job with such mundane tasks as password security than the great unwashed masses.  However, time and time again, this turns out not to be correct.  Yesterday, Twitter co-founder Biz Stone, posting in the company’s blog, revealed that a hacker had broken into an employee’s personal e-mail account and then gained access to that employee’s Google Apps account, which contained “notes, spreadsheets, ideas, financial details” – well, you get the picture.

Although Stone tries to emphasize that this has nothing to do with any vulnerabilities in Google Apps per se, the very fact that anyone can log into a Google Apps account from any browser if you have the correct user name and password does increase a company’s exposure.  Companies that keep their confidential information behind a corporate firewall in systems such as Lotus Notes or Microsoft SharePoint, are indeed less vulnerable simply because their systems could not be hacked with just a simple user name and password.

Multiple studies have revealed that close to half of computer users tend to use the same password over and over again – typically with the same, easy to remember, user name.  Indeed, TechCrunch, a blog that received Twitter’s confidential documents from the hacker, reported that Twitter uses the password “password” for its servers (presumably, it’s been changed by now).  The same article revealed that Twitter had also used a co-founder’s first name, Jack, as a user name for servers.

Moral of the story: use complex passwords with numbers and symbols interspersed.  Do not use words found in a dictionary.  Even better: use passphrases, i.e. concatenated words such as “thisismypassphrase123″.  Use a different user name/password combination for each account.  If one account is hacked, this will ensure that your other accounts remain safe.  Finally, do not leave passwords visibly written down.  Believe it or not,  I still see Post-It notes with passwords attached to monitors when visiting other companies.

David M. Goldes is the president of Basex.

Nortel to Liquidate Assets, Sell Wireless Unit to Nokia Siemens

Friday, June 19th, 2009 by Jonathan Spira

Nortel Networks, one of the oldest companies in the tech sector, will liquidate assets and sell its CDMA and LTE Access units to Nokia Siemens for $650 million.  Nokia Siemens is a joint venture of Nokia of Finland, a company founded in 1865, and Germany’s Siemens, founded in 1847.   The joint venture is acquiring a unit that manufactures CDMA mobile technology, a system used by several U.S.-based mobile operators including Verizon Wireless and Sprint.  The deal gives Nokia Siemens a significant customer base in the U.S., something that has eluded the company up until now.

Nortel said it was in advanced talks to sell other businesses and that it was applying to have its shares delisted from the Toronto Stock Exchange effective June 22.  Nortel filed for bankruptcy-court protection in January and has been attempting to sell its assets since then, having amassed almost $7 billion in losses over two years.

The company has a colorful history dating back to its founding in 1895 as Northern Electric and Manufacturing, a supplier of phones and other devices spun off from Bell Telephone of Canada.  It started looking into ways of using fiber optic cable in the 1960s at which time it also began designing digital telecommunications equipment.

In 1976, the company changed its name to Northern Telecom and announced Digital World, a family of industry-leading digital telecommunications products.  The DMS-100 became a mainstay of telephone company central offices (it could handle 100,000 subscriber lines without breaking a sweat) and the DMS line contributed greatly to the company’s profits for 15 years.

In 1998, with the acquisition of Bay Networks, the company changed its name once again, this time to Nortel Networks.  It gained prominence in the late 1990s as a manufacturer of fiber optic gear used to transport massive amounts of data over the Internet but was also one of the first casualties when the telecom bubble of the time burst, sending the company’s market capitalization from $398 billion (Canadian) in September 2000 to $5 billion in August 2002.

Now the company, which dropped “Network” from its brand but not from its legal name, will fade from history following a decade of failed and expensive acquisitions.  Its strategy, which was to buy established companies, didn’t work, largely due to an inability to integrate products from the newly-acquired entities into a common vision.  In addition to the Bay Networks acquisition for $9.1 billion, the company acquired software maker Clarify and then Alteon Web Systems for a total of $10 billion in 1999 and 2000 respectively. The company’s buying spree continued with Cambrian Systems for $300 million, Shasta Networks for $340 million, all the way to DiamondWare (3-D stereo conferencing) and Pingtel (SIP software) in 2008.

Nortel’s strategy contrasted greatly with that of companies such as Cisco, which only acquired small and innovative companies at much lower cost and then successfully integrated them into the business.  Today’s Nortel is also a stark contrast to Northern Telecom ca. 1976, when it announced Digital World.

As competition has intensified from North American, European, and low-cost Asian rivals, among them Alcatel-Lucent and Huawei Technologies, the company’s shares sank into the penny range in the months prior to its bankruptcy filing.  In addition, the global economic crisis has slowed spending on the gear that Nortel offers (which includes equipment for the enterprise as well as for telephone companies).

A likely acquirer for Nortel’s enterprise unit is Siemens Enterprise Communications, which has been in growth mode since it became a joint venture between Siemens AG and the Gores Group.

Jonathan B. Spira is the CEO and Chief Analyst at Basex.

FiOS Follies

Wednesday, February 18th, 2009 by Jonathan Spira

First announced in July 2004, Verizon FiOS couldn’t come to my neighborhood in New York City soon enough. Using fiber-optic connections instead of copper wire to bring telephone service, Internet, and television into the home, FiOS (which stands for Fiber Optic Service) was certainly worth the wait. So was the pain of the installation process and problem solving that followed.

After five hours plus, and a call for a more experienced installer, my FiOS service was up and running – more or less.

The installation consists of bringing the fiber-optic connection into the home and terminating it in an optical network terminal (ONT), which serves as an interface to inside wiring for telephone, television, and Internet access.

The TV service itself is superb, with better picture quality than our cable company (Time-Warner) had ever provided. The multi-room DVR (digital video recorder) system allows streaming of recorded programs (HD and standard) to other TVs in the home. Widgets provide local traffic and weather and local and national news on the top of the screen while programs continue in a slightly smaller size below.

The FiOS Interactive Media Guide has an easy-to-use tabbed interface and allows searching for words that appear anywhere in the description. One can remotely program the DVR via the Web (or using a Verizon mobile phone). The service features over 100 HD channels, 500 all-digital channels, and 14,000 video-on-demand titles (8,500 are free).

The Internet service is lightning fast. It consistently measures close to 20 Mbps, about seven times faster than my DSL service ever was. It’s so fast that my partner and I can each watch a different streaming TV show on our respective computers without any problem (with DSL, one show was frequently more than the service could handle).

It was the plain, old telephone service (known in the industry as “POTS”) that turned out to be the big problem. The day after installation, I noticed that many of my calls were not going through; instead, after dialing, I would hear an ACB recording (“We’re sorry, all circuits are busy…”). After weeks of investigation, this turned out to be a software error; my phone line was coded as an account disconnected for non-payment. I also found that I couldn’t place a call a few times a day; pressing the number pad would simply not break the dial tone. Then a reorder tone (sounds like a fast busy signal) would follow, then a message stating “if you’d like to make a call, please hang up and try again.” I told the repair bureau it was a bad line card but they didn’t seem to believe me. This problem took over two months to resolve and involved dozens of phone calls and the implementation of odd fixes at the phone company’s suggestion (twice they had me unplug all of my phones and they replaced the ONT and also sent a technician to check the inside wiring). Two months later, the problem was determined to be a bad line card in the Nortel softswitch.

A few small glitches remain to date. The remote set-top box loses the connection to the main DVR several times a day and it also has trouble playing recorded programs longer than 30 minutes. In such cases, it loses track of where it is. (Verizon promises a fix for the first problem shortly and advises that the second problem is being worked on.) In addition, the problem in placing a call mysteriously returned for two days recently and then disappeared again.

By this time, you are probably wondering if getting FiOS is worth it – and my answer is a resounding “yes.”

The clear sharp television picture and the lightning fast Internet connectivity are simply head-and-shoulders above any other service I have seen and I saved the best for last. Even with faster speed and sharper picture, I’m saving money. A bundle including TV, Internet plus telephone service is $99.99 per month plus taxes and fees (previously I was paying 60% more for inferior service).

Jonathan B. Spira is the CEO and Chief Analyst at Basex.

Microsoft Turns Blue

Friday, October 24th, 2008 by David Goldes

Is Microsoft feeling blue, taking a subtle swipe at Big Blue, or seeing blue skies ahead?  On Monday, Microsoft proclaimed a possible third era of operating systems with the introduction of Windows Azure, a cloud operating system, essentially an OS that exists within a network framework.

Azure manages the relationship between software residing on computers and the Web, where data and services may reside.  This paves the way for a wave of applications that support a broad range of devices but it can also be viewed as yet another cloud computing initiative by a major tech company.  It also puts Google and Amazon.com into even more direct competition with Microsoft.

Windows Azure is part of the Azure Services platform, a development platform for applications that will use the service.  By integrating Windows Azure with programming tools such as Visual Studio 2008, Microsoft is making it easy for developers to create applications for the new environment.

Microsoft unveiled Windows Azure at the Microsoft Professional Developers Conference this week because it needs to win this community over; without a corps of developers creating compelling applications, customers will go elsewhere.  The company faces some additional obstacles, not the least of which is whether companies will trust it sufficiently with proprietary data and applications.

David M. Goldes is the president of Basex.

Did 1997 Stand Still? Looking Forward While Looking Back

Tuesday, January 15th, 2008 by Jonathan Spira

In preparing our annual State of the Technology Industries diatribe, I thought to look back at what I penned one year ago.  It seems that, in many respects, time stood still.  Imposssible, you say.  The industry now moves on Internet Time.  Let me clarify:  the industry did indeed make tremendous progress.  But that seems to have occurred despite itself.  This year, we are changing our perspective: rather than go company by company, we look to the The Desktop to ease us into the year.  In a related article, we’ll review our 1997 musings and see how things panned out.

The desktop computer is perhaps the most important interface that one has, and will continue to have, over the next few years.  Despite promises of body-networks and the like, the average computer user will sit himself before a computer workstation and be saddled with its brief but illustrious user interface history.  With the court’s permission, 1998 should see the advent of Windows 98 and NT 5.0.  For starters, limited self-healing features and self-administration with the attraction of multi-user Windows may make for a more centralized and more easily managed environment (read: significantly lower costs for both hardware maintenance and support).  Both OSes will benefit from a common addressing scheme for hardware, the Windows Driver Model, making upgrades from Windows 98 to NT 5.0 a possibility.  [The most recent Windows 98 beta that arrived at my desk hinted that it has a Windows 3.x migration facility; unfortunately, I don't have any Windows 3.x machines left to try this out on that would support Windows 98's hardware requirements).

1998 will also be a make-or-break year for Java, which is the apparent main alternative to Windows.  A new Java Development Kit (JDK), combined with significant recent product introductions showing a committment to the platform (Lotus' eSuite for one) may signify that Java is now ready for prime time.

For Apple, 1998 may be a true crossroads.  Rhapsody, its multithreaded, multi-tasking, cross-platform OS, will offer developers yet a third alternative to Windows.  Although Rhapsody does not pretend to achieve Java universality, its ability to run on Intel, PowerPC and Alpha will be a plus, since its development environment also runs under NT.  Will the wait Be worth it?  Or will Mac users flock to Be insead of waiting?

Will the hardware be up to the challenge?  Memory prices continue to decrease dramatically, making the possibility of 128 Mbyte workstations reality for some.  [Of course software developers are aware of this and already seem to be increasing minimum memory requirements without recognition that the installed user base of machines has yet to upgrade to 32 Mbytes.]  Intel continues to offer its Motherboard-of-the Month Club option and one of these days the underlying architecture will catch up to the speed of today’s supersonic microprocessors [didn't we say that when 80386's were introduced as well?].

1998 should also be a do-or-die year for diskless network computers and their variants.  Will these turn out to be cheaper and more easily managed the their larger PC cousins?  Or will they turn out to only be a (significant) 3270 upgrade path?

So what’s missing in the Desktop Environment.  Ah, yes.  Bandwidth.  Coming this year:
Gigabit Ethernet at a pricepoint that is implementable
Rotorouter for the Router – better intelligent switching to eliminate bottlenecks
One agreed-upon standard for 56 Kbps modems (ok, this is wishful thinking but we better agree on something before 56 becomes too slow)


1998 should be the year of what we at The Basex Group have named the “Virtual Corporate Community”.  Wipe that smirk off your face:  the picture you should have in your mind right now is not a bunch of funny-looking avatars lurking in The Palace talking about politics or sex.   To learn more about this phenomenon, please click here.

Did 1997 really stand still?  Of course not.  1997 was a year for clarification and regrouping.  But the future picture is much brighter.

Jonathan B. Spira is the CEO and Chief Analyst at Basex.  This article originally appeared in the Basex Online Journal of Industry and Commerce (BOJIC).

Mini Y2K Looms in Daylight Saving Change

Tuesday, March 6th, 2007 by Jonathan Spira

Spring Ahead with Caution

This coming Sunday morning, while most people are asleep, the United States and parts of Canada will switch to Daylight Saving Time at 2 h local time.  This is in accordance with the Energy Policy Act of 2005 and is three weeks earlier than in previous years (it ends one week later than usual, on the first Sunday in November).  If you don’t think that these changes are a big deal, change the time on your servers by an hour and see what happens.  Its impact will extend well beyond computers, to legions of business travelers and mobile knowledge workers, among others.

The change means the United States will be out of sync with the rest of the world for longer than usual.  Europe used to change its clocks one week before the U.S.  Now most of Europe will switch to daylight saving time on March 25, two weeks later.  [Most of Asia, Africa, and South America do not observe daylight saving time at all.]

Daylight Saving Time is a system of managing the changing amounts of daylight that occur during the year, with a goal of maximizing daylight hours during typical waking hours.  It was first proposed by Benjamin Franklin in 1784 but was not broadly adopted until the early twentieth century.  By adjusting clocks ahead by an hour, people can have more daylight available during the workday.  For example, in the case of someone who typically awakens at 7 h, since in the spring the sun rises earlier each day, an individual would have to wake up at 6 h to take advantage of the additional daylight.  Instead, by moving the clock ahead by one hour, that person can continue to wake up at 7 h and enjoy more daylight in the evening hours.

The last change to the Daylight Saving Time schedule was in 1986, when legislation changing Daylight Saving Time from the last Sunday in April to the first Sunday in April was enacted.

In case you are wondering why we are doing this now, it’s not just to enjoy a little bit more sunlight.  Although the savings are almost minuscule on an individual basis, the American Council for an Energy-Efficient Economy, a non-profit group, estimated that the cumulative benefit of the change through the year 2020 will be a savings of ca. $4.4 billion and 10.8 million metric tons less carbon sent into the environment.  According to the U.S. Department of Transportation, for every day we are on Daylight Saving Time, we trim one percent of the country’s electrical consumption.

Since late last year, companies have sent out all-hands memos to employees asking them to help identify systems that might be impacted by the time change.  These systems range from automated wake-up systems in hotels to systems that schedule airline crew members and slot aircraft for gates.  In addition, many computer-to-computer systems may also be impacted.

Most knowledge workers should be covered by now, at least insofar as their desktop or laptop computers are concerned.  Microsoft released a single global time zone update for Windows XP Service Pack 2 and Windows Server 2003 (and for Windows Server 2003 Service Pack 1) that automatically installed.  This update included updates for all DST-related changes from 2007 or that have taken place since the operating system’s original release.  The updated time zone definitions will also ship with Windows Vista.  Windows XP SP1 and older operating systems have passed their end of support dates and did not receive the update although they can be manually updated in some cases.

That covers operating systems but doesn’t mean that we are out of the woods yet.  Software that supports calendar entries, such as Lotus Notes and Domino and Microsoft Outlook and Exchange, and other time-zone-aware calendar systems, are also impacted.  If you created an entry that falls within the extended Daylight Saving Time period before the application of extended DST rules, it will appear incorrectly after the extended DST rules have been applied.  Entries that fall between March 11, 2007 and March 31, 2007 will appear an hour later than originally scheduled.  Entries that fall between October 28, 2007 and November 4, 2007 will appear an hour earlier than scheduled.  For example, if you create an entry today for 10 h on Monday, March 12, 2007, the entry will display at 11 h after the rules are applied.

What you can do to avoid problems:

First, double check any calendar entries or plans for the period March 11 -  March 31, 2007 and October 28 – November 4, 2007.

Second, make certain to adjust or update your operating system to apply the changed Daylight Saving Times rules if this hasn’t already taken place.  PDAs such as Palm Treos or BlackBerry devices should also be updated.

Third, remember that Daylight Saving Time is not observed in Hawaii, American Samoa, Guam, Puerto Rico, the Virgin Islands, and Arizona (with the exception of the Navajo Nation).  Until 2006, the counties in the Eastern Time Zone of Indiana did not observe Daylight Saving Time and remained on standard time year round.  As of April 2006, all of Indiana observes Daylight Saving Time.

Finally, get a good night’s sleep.

Jonathan B. Spira is CEO and Chief Analyst at Basex.

Inventing the Internet

Saturday, January 3rd, 1998 by Jonathan Spira

Where Wizards Stay Up Late: The Origins of the Internet, by Katie Hafner and Matthew Lyon, published by Simon & Schuster.

In the late 1960s, the Defense Department’s Advanced Research Projects Agency (ARPA) funded the creation of a network to facilitate computer communication between DOD-funded research universities.  This experiment became the basis of a facility we take for granted everyday, the Internet.

Where Wizards Stay Up Late narrates the story of the small group of computer scientists, researchers and engineers who, both independently and working in small groups, studied and tested just what it would take to link computers spread across an entire nation.

Wizards tells the story of Bolt Beranek and Newman (BBN), a small consulting firm in Cambridge, Massachusetts, who first built the “glue” that held the network together, the IMP (Interface Message Processor), and managed network operations for years to come.

Wizards tells the story of how the first Request for Comment (“RFC”) was created (in a bathroom), and how, through what one might refer to as strategic stumbling, a series of discoveries, from electronic mail (then “network mail”) to what has perhaps defined Internet iconography more than anything else, the “@” symbol.

By the time the ARPANET was retired in 1990, it had become fully integrated in its descendants, the NSFNet and the Internet itself.

Wizards is written in a lively and totally comprehensible manner; along the way, its compelling story answers many questions you may or should have had about why things exist on the Net the way they do.

And I stayed up late just to finish this book.

You can order Wizards now from Amazon.com.

Jonathan B. Spira is the CEO and Chief Analyst at Basex.  This article originally appeared in the Basex Online Journal of Industry and Commerce (BOJIC).

Moving Along The I-Way

Sunday, February 2nd, 1997 by Jonathan Spira

In December 1996, the FCC refused to allow phone companies to charge a different rate for data calls than for voice.   Since the divestiture of AT&T in 1983, under the Modified Final Judgment (MFJ) that has been supervised by Judge Harold Green, long-distance phone firms pay local phone companies for access to local phone lines, but Internet-access firms and other non-voice service providers are exempted from such charges.

With the exploding popularity of the Net, local telephone companies are charging that this exemption has left them unable to unclog the public switched voice networks and, more significantly, has discouraged them from making the investment required in their data networks that would truly let the Internet meet consumers’ needs.

The phone companies argue that the small number of users should bear the cost of upgrading and supporting access to the Net.  Internet Service Providers (ISP’s) counter-argue that access charges would slow or possibly even strangle the growth of the Net and that the phone companies are crying wolf because they are trying to get into the Net access business themselves.

Blair Levin, the chief of staff for FCC Chairman Reed Hundt, spoke at the Comnet and noted that “Now is not the time to act,” indicating that the federal government is unlikely to subject Internet providers to access charges for using local telephone lines.

Levin also mentioned that it would be unreasonable to impose old rules (access charges) on new technologies and that reform was needed on the part of the FCC and access charges.

The underlying logic of the phone companies is that while typical residential voice calls average 5 (3 ccs) minutes in duration (typical business calls use twice as much line time), data calls can go on for hours, overtaxing the public switched network. They want these data users to pay for upgrading the network.  It is perhaps a tribute to the increasing maturity of the Internet in terms of richness and variety of content that more and more users are staying on line longer.

This increase in data calls does appear to have been a boon for the local exchange carriers, in that second residential line installations were up 25% in 1996, compared to an average increase of 8% in years past.

Curiously enough, NYNEX, in the New York City metropolitan area, created a new twist on unlimited, untimed service last year. [Previously, in New York City, the only calling plan available was one where callers paid for each local call.] They offered an unlimited service to residential customers, based on call usage patterns of that customer in a given three-month period that rolled forward depending on when you signed up.  It seems paradoxical for them to argue that they’re losing money on local calls while they’re creating unlimited services at the same time.

By any measure, however, the increase in volume and length of Internet-related calls does have an impact on the existing circuit-switched voice telephony networks.  The potential for bottlenecks is greatest at the terminating central office (CO) switch that provides service to ISPs, but they can also occur at originating points where there is a high concentration of Internet users or telecommuters.

Depending on the extent of congestion, users can expect delayed dial tone, denial of dialtone, and ACB (All Circuits Busy) messages — clearly an unacceptable level of service in an era of instant gratification.

What is the alternative?  Nortel has developed an innovative approach, Internet Thruway, which minimizes traffic congestion by migrating long-duration data and Internet calls from the public switched network onto a more efficient, highly scalable packet-data network.

This is technically an enhancement to the public switched network, in that it maintains the traditional circuit-switched pathways for voice calls while adding packet-switched capabilities for data.  The telephone companies will enjoy a new revenue opportunity by providing outsourcing of modem management functions to the ISP, removing obstacles presently inherent in the ISP’s business model.

As the popularity of the Internet continues to soar, more and more telephone company customers will go “on line.”  It is likely that many of these customers will use phone company facilities for access.  As this occurs, it will no longer remain a question of how to service an “elite” minority of users but how to provide the ubiquitous “data dial tone” that has long eluded the public-at-large.

Jonathan B. Spira is the CEO and Chief Analyst at Basex.  This article originally appeared in the Basex Online Journal of Industry and Commerce (BOJIC).

The State of the Technology Industries January 1997

Wednesday, January 15th, 1997 by Jonathan Spira

As industry analysts, we follow the comings and goings, trials and tribulations of dozens of companies who impact the industries we closely follow.  Their fortunes rise and fall based on how quickly they well they are able to maintain mindshare as well as marketshare.  Unlike financial analysts, who apply their own proprietary models to examine (exhaustively, I might add) balance sheets, filings, ROI – and sometimes even have more accurate information than the data provided by the companies themselves, we  look at companies from a far more subjective
perspective:  innovation counts, as does mindshare, as does the ability to turn around a company’s focus on a dime.

Funny things have happened from our musings.  Last year, MSNBC liked a sentence that our own David Goldes penned so much that they asked for (and received) permission to quote it.  Next thing we knew, that line (along with Goldes’ e-mail address) was appearing in large ads in the Wall Street Journal and in bus- and train-length ads on the sides of moving vehicles.

There are far more companies in the technology sector than we could hope to follow and get to know; therefore, inclusion in our list is perhaps indicative of our own leanings as much as it was the ability of  their marketing and analyst relations people to keep their company’s name in front of us.

This review, by the way, is presented perhaps more as a summary of the past year rather than a harbinger of the future; past performance is no guarantee as they say.  And now, the envelope please:

3Com.  Even its basic producs (network cards) easily outdistance the competition.  Needs to align its acquisitions and make everything work together.  It’s harder than it sounds.  How about a stadium on the east coast?

AOL.  Compuserve.  Prodigy.  Dramatic changes in the market (esp. pricing) can be devastating.  So are busy signals for end users.  AOL’s restatement of its financials proved it lost 17 cents on every dollar it earned.  CompuServe loses 10,000 members a day (and is not adding nearly so many new ones).  GEnie has 20,000 members left.  And Apple logged eWorld off.  These companies may become footnotes before they know it, if they don’t leverage their base as an ISP successfully.

Apple.  NextStep and Jobs are an unbeatable combination, but the OS doesn’t work on Apple’s Mac platform.  Will loyalists wait until ’98?

AT&T.  They started the year by announcing the spin-off of NCR and Lucent.  Now, with John Walter, they hope to concentrate on their core markets.  I hope they find them.  A further downsized AT&T, concentrating on the modern day equivalent of “long lines,” is a powerful contender; adding new media (read: Net access) into the mix while leveraging its customer base is a potentially lethal combination.

Cisco.  Its products run the Net.  ‘Nuf said.

Computer Associates International.  As it continues to integrate its acquisitions (most recently Cheyenne), it has become almost as beloved as Microsoft in its competitors’ hearts.

Corel.  Only Michael Cowpland could save WordPerfect.  Applying the Corel marketing machine to the product is its only hope (along with restoring the expected high level of tech support).  Corel also understands the WordPerfect legacy user.  High expectations here.

IBM.  Gerstner has reenergized and refocused the company (and Wall Street agrees).  Its mainframe business has never been hotter.  Whither the mini-computers?  Whither OS/2?

Intel.  Has learned the art of eating its own children and then growing new (and faster) ones.  And Intel continues to push the envelope of the industry with a basic ingredient: continuing, genuine innovation.

Lotus.  Domino could dominate, but the user community needs to understand its significance first.  By unbundling the unique aspects of Notes (replication, for example), it can create a new standard, but risks weakening the growth of the overall Notes market.  For those who doubted, the wisdom of IBM’s investment is definitely beginning to pay off.

Lucent.  No longer tied to AT&T, they have created tremendous mindshare and an ability to sell to what formerly was Lucent’s AT&T-parent competition.

Microsoft.  NT Advanced Server flies.  Windows 95 has the workstation mindshare.  If only you could upgrade Win95 to Windows NT workstation.

MSNBC.  Having trouble bridging news and on-line functionality; possibly due to huge rift between two corporate cultures.  It’s an idea whose time has come – all they have to do is innovate rather than repurpose.

Netscape.  Refocusing, but de-emphasizing the Navigator name which owns the browser mindshare market.  Now aiming at Lotus but has yet to fire more than a shot across its bow. Missing ingredient: tools and a datastore

Novell.  Owned mindshare for networking, but Microsoft caught on.  Needs to leverage its roots.  Divesting non-core products and non-core execs has been a good move.  IntraNetWare makes the Net a plugin to Novell shops.  The core product needs to be marketed to non-IS managers de-emphasizing speeds and feeds.

NCR.  Strong brand name returns; hopefully not too much damage was inflicted.  Needs to return to roots in banking and retail systems.

Oracle.  Now that its own internal financial reporting has been straightened out, its ownership of the database market is unquestioned.  Other companies have learned that moving away from core competencies can be devastating?  Does Oracle envision a world of Oracle hardware (NCs) running Oracle software?

SGI.  Why Cray?

Sun Microsystems.  JAVA and JAVAStation.  Need we say more?  Don’t lose the momentum here.

Symantec.  Web, JAVA, Act, and Peter Norton.  Neat package.  Neat integration.

Tandy.  Needs to go back to its Radio Shack roots.  In its day, the TRS-80 was considered a market leader.  Now they are narrowing their Universes and abandoning their (Computer) Cities.

Wang.  No hardware, no debt.  Its workflow and imaging innovations are set to propel it forward.

Jonathan B. Spira is the CEO and Chief Analyst at Basex.  This article originally appeared in the Basex Online Journal of Industry and Commerce (BOJIC).