USNO Y2k Bug Hits Headlines
10 January 2000
Widely recognized as "America's timekeepers",
the US Naval Observatory suffered the ignominy of, perhaps, the
most embarrassing Y2k slip-up reported during the first days of
this new century. Their website's millennium countdown clock seemed
to be suffering the same over-indulgence that afflicted many of
its countrymen during the period, as it precisely (but inaccurately)
indicated that there were almost 700 thousand days until the year
2000 !

Time Bombshell : Leap Second Re-think
12 January 2000
Global timekeeping gurus are considering proposals
to discontinue "leap seconds" which are periodically,
but irregularly, introduced into Coordinated Universal Time (UTC).
The arguments for scrapping the practice include the fact that many
high-tech navigation systems can't handle leap seconds very well.
GLONASS, the Russian equivalent of the American GPS, goes off-line
for leap second adjustments. Also, problems can occur in interfacing
between systems that handle leap seconds differently and human error
- whether it's to be added or subtracted - is a real possibility.
These days, of course, the ubiquitous Internet also figures in the
justification in terms of network synchronization. The Network Time
Protocol (NTP) uses the Internet to transfer time which was particularly
on the mind of one corespondent to the USNO who opined:
"If leap seconds went away, the NTP community would
worship the ground you walk on. Leap seconds introduce a manual
discontinuity in the NTP time scale. It takes a while to propagate
leap seconds through the hierarchy. They are a tremendous headache
in the NTP world because they cannot be predicted. I think it
is a very true statement that all GPS users would vote against
continuing leap seconds, not just NTP users. Many telecommunications
circuits use GPS or atomic clocks to keep cellular phones in operation,
and leap seconds are a nuisance to them as well."
Those against the move, perhaps somewhat tongue-in-check, argue
about the impact upon society if our civil timescales get progressively
out of sync with timescales linked to our planet's physical characteristics.
But the scale of this problem is only, perhaps, a minute over the
next century; we certainly won't find it dark at mid-day at Greenwich
(UK) !
Amongst the possible alternatives to scrapping leap seconds altogether
would be to insert larger discontinuities less often, inserting
at predefined times, simply using international atomic time (TAI)
or, the more heretical suggestion of redefining the length of the
second.
Feedback on the subject is currently being sought through a USNO/IERS
survey but don't worry about any "surprise" decisions;
before the relevant international bodies make any changes, full
debate and publicity is promised.... So watch this space !
What's a Leap Second, Anyway?
Since the early 1970's these have been used to adjust UTC so that
it's consistent with the Earth's rotation as determined by astronomical
observation (Mean Solar Time). UTC is the basis of every nation's
civil time and is maintained using a weighted-average of the performance
of hundreds of atomic clocks around the world. The US Naval Observatory
provides further information about leap seconds.
ISO/IEC17025 Published.... Finally !
14 February 2000
Hurry, hurry, hurry... they're selling like hot
cakes !
It's taken several years in the making but, at last, the eagerly
anticipated replacement to ISO/IEC Guide 25:1990 has now been published.
This new 26-page standard, entitled General Requirements for
the Competence of Testing and Calibration Laboratories, is
available directly from ISO or national document distribution companies.
Far more extensive than the now obsolete Guide 25, 17025
incorporates the quality management criteria of the 1994 versions
of ISO9001/2 as they apply to laboratories. Which is ironic, of
course, because this standard is being significantly revised and
should be published later this year.
Metric Only Delayed Again
...while Europe Split on SI Units ?...(maybe !)
22 February 2000
Two measurement reporting practices will be adopted
in the European Community if publications on the official Europa
website are to be believed.
The Official Journal of the European Communities L34 Volume
43 dated 9 February 2000 reports on the mandatory legislation "Directive
1999/103/EC of the European Parliament and of the Council of 24
January 2000 amending Council Directive 80/181/EEC on the approximation
of the laws of the Member States relating to units of measurement".
The English language version of the Directive demands that the
kilo prefix symbol shall become "K" and the hecto prefix
become "H". Until now those symbols have been reserved
for the exclusive use of Kelvin (temperature) and Henry (inductance).
Apparently, the French and Greek concur with this practice according
to their versions of the Directive. Those three allies are in the
minority though; Germany, Sweden, Finland, Italy, Spain, Holland
and Portugal will continue to use the traditional "k"
and "h" SI multiplier symbols.
A spokesman denied that this will lead to any greater confusion
in the marketplace, observing "We're striving for a federal
Europe and the value of diversity must, therefore, be upheld."
European standardization at its best !
Meanwhile, on a more serious note, the Directive confirms that
products labeled in non-metric and metric units (so-called
dual labeling using "supplementary indications in non-legal
units") will continue to be authorized for sale in the Community
for a further, though undefined, period. An earlier extension
to the metric-only deadline established in Directive 80/181/EEC,
which prohibits use of English (Imperial) units, expired at the
end of last year. Vive la difference, as they say in France.
Editor's Note
Most of the above article is written "tongue-in-cheek"
because we can safely assume the English, French and Greek documents
to be in error regarding the SI amendments. It just goes to show
what can happen when lawyers get involved with metrology though
!
More Metric Mayhem
...or When is a Quarter-pound not a Quarterpounder ?
...or Do you want fries with that ?
10 March 2000
Following adoption of the European Union's metrication
law in January, stories regularly appear in British newspapers that
highlight, possibly, the unforeseen impact of the legislation on
British culture, as well as the above conundrum.
Required to enforce the law demanding loose food to be sold using
only metric units, trading standards officials in the north of England
refused to bend the regulations for a museum's candy shop... even
though the Beamish Museum is a "living" recreation of
the way people lived at the turn of the last century (circa 1900).
Managers were ordered to cease selling such local favorites as black
bullets and cinder toffee by the traditional quarter-pound in favor
of grams. The argument that this would be ridiculous in a replica
of an early twentieth-century tuck shop was rejected. But maybe
the museum's proposition would have had been more favorably received
if they charged 1900 prices ?
Given the British Weights and Measures Association's stance on
the matter, their director gave a predictably supportive quote:
"It's a pity public money is being squandered to wreck the
unique heritage of our nation."
Meanwhile, fast-food giant McDonald's is keeping its quarter-pounder
on the menu despite the metric laws. Officially, the legislation
means it must now be sold as a "113-grammer", but their
spokesman rejected the suggestion of a new name appearing on the
menu. "Quarterpounder is a registered trademark name",
he asserted, "not an indication of weight."
Free Uncertainty Aid
21 March 2000
The European co-operation for Accreditation has
released a new document on calibration uncertainty (EA-4/02). It
offers step-by-step guidance and is supported by a good number of
examples. Refer to our Uncertainty Resources page for the
opportunity to view / download.
GPS SA Free
17 May 2000
In an unexpected move that shook many users of
the US Defense Department's Global Positioning System, President
Clinton announced on May 1st that "selective
availability" would be turned-off at midnight that very day.
SA is, essentially, interference added to the transmitted signals
whose effect is to reduce receivers' location-finding ability.
More information about the President's decision and GPS in general
is available from the Interagency GPS Executive Board.
Other than for navigational purposes, GPS is also used for timekeeping
and standard frequency dissemination and these applications are
also expected to benefit from SA removal. Several of Agilent's calibration
centers around the world use GPS-disciplined oscillators to monitor
the performance of their own atomic frequency standards -- we'll
keep you posted as to our experiences of the new, noise-free GPS
!
World Metrology Day
...the award for the Best Kept Secret
20 May 2000
A decision was taken during the twenty-first
General Conference on Weights and Measures (CGPM) held last October
in Paris, France to annually mark May 20th as World
Metrology Day.
The date was chosen because
it marks the signing of the Metre Convention in 1875. This Treaty
was initially signed by 17 nations with the aim of providing a formal
basis for international agreement on units of measurement based
on the metric system. There are now 48 Member States.
It's interesting to note that the CGPM chose to
make World Metrology Day a "fixed feast" falling on the
same date each year (like Christmas Day). This contrasts with the
less precise World Standards Day which ISO deems to be a
"movable feast" (like Easter) and, indeed, the exact date
of which can vary from country to country.
This year, World Standards Day will be celebrated
on October 18th in the USA.
The UK NPL have compiled a calendar of worldwide
events to celebrate World Metrology Day.
ISO9000:2000 Moves to Final Stage
01 August 2000
In May, ISO announced that its membership had
returned a positive vote on the Draft International Standard version
of the ISO9000 revisions in development. The comments received with
the votes were subsequently considered by Technical Committee 176
at their meeting last month in Japan. Agreement was reached on revisions
to the text and that the document should go forward to the final
stage in the process -- publication of a Final Draft International
Standard.
The FDIS versions of ISO9000, ISO9001 and ISO9004
are expected to be published in September. The organization's national
member institutes are allowed a two-month period to cast their votes
which, assuming three-quarters of votes are in favor, should mean
that the updated standard will be released by the end of the year.
Unaccredited Calibration Acceptable
15 August 2000
The British accreditation agency (UKAS) recently
held a series of seminars to explain how its adoption of ISO17025
will affect accredited laboratories. At the London meeting, useful
clarifications were gained by the assembled laboratory representatives
but one discussion point in particular was, potentially, very significant.
During its development stages people will have scrutinized its clauses
but many may not have fully appreciated the impact of paragraph
5-6-2-1-1 so here's the scoop...
UKAS confirmed that they will not expand upon
17025's criteria; several of their current accreditation documents
will be SCRAPPED and the remainder revised as GUIDANCE only. Not
really much of a surprise until attention was drawn to the effect
of 5-6-2-1-1 which is that UKAS will no longer require traceability
to be obtained from another accredited laboratory. Assessors will
accept non-accredited calibration certificates provided that sufficient
evidence is available to confirm that the potential supplier complies
with the standard and that the (UKAS) auditee has the critical technical
competence to make such an assessment.
The question of whether the other accreditation
agencies have the same attitude to the standard (or will maintain
additional criteria) was raised but not definitively answered. However,
attention was drawn to the fact that some customer standards (e.g.
the automotive industry's QS9000) stipulate accredited suppliers
must be used and that some countries or industry sectors impose
legal obligations for accredited certificates. Taking-up the opportunity
presented by this radical policy change could open up the whole,
proverbial "can of worms".
The presentation material is available from the
UKAS website, including notes from the most useful "Question
& Answers" (Q&A) sessions.
Time Lord Honored
13 September 2000
The
cult television show Doctor Who has a new challenger to play
the role of the Universe's principal time lord.... Agilent Labs
atomic clock pioneer Len Cutler (pictured right) was recently named
Inventor of the Week by the Massachusetts Institute of Technology.
This prestigious honor was bestowed under the
Lemelson-MIT Awards Program, part of the Lemelson National Program
in Invention, Innovation and Creativity. The program's mission is
to inspire a new generation of American scientists, engineers and
entrepreneurs by celebrating living role models in these fields.
The first atomic clock was invented in 1948. In
the early 1960s, HP set out to design a cesium clock for commercial
purposes. In 1964, Cutler and his colleague Al Bagley succeeded,
inventing the 5060A Cesium Beam Clock. This was the first all-solid-state
cesium-beam chronometer, whose frequency standard was soon adopted
by the US National Institute of Standards and Technology and other
scientific centers around the world. Cutler's 1964 clock coordinated
international time to within a microsecond, whereas previous efforts
had pared accuracy down to only a millisecond. Cutler continued
refining both the design and the constituent parts of cesium clocks,
as well as researching in quantum mechanics. His most recent triumph,
in 1991, was the 5071A: at twice the accuracy of its predecessor,
it remains the world's most accurate commercial clock, losing about
one second every 1.6 million years. 5071A's also account for 82%
of the data relied on to keep the International Atomic Time Standard.
World Standards Day
11 October 2000
Initially celebrated thirty years ago, the original
objective of World Standards Day was to recognize the birth of ISO
(the International Organization for Standardization). ISO's inaugural
meeting took place in London, United Kingdom on 14th October 1946. Now, the aim is to raise awareness of the importance
of global standardization to the world economy and to pay tribute
to the thousands of volunteers who contribute their expertise in
committees and work-groups.
October 14th is the formal date of
commemoration but, unlike World Metrology Day, the celebration of
World Standards Day occurs on different dates in various countries.
The USA has denoted October 18th this year but there
will be party balloons all week as many special events around the
day are planned. Contact your own country's national standards body
or metrology institute for details of how you can join in the fun
locally !
At Last ! Accreditation Gets a Global "Thumbs-Up"
15 November 2000
It's taken a quarter of
a century but the first pan-world mutual recognition agreement for
accredited calibration and testing has been brokered by the International
Laboratory Accreditation Cooperation at its recent conference in
Washington DC.
An insistence that products
be retested at importation to check conformance with local standards
and regulations, was identified in the mid-1970's by the World Trade
Organization as a widespread technical barrier to trade that adds
delay and cost to making goods available to a wider market.
Coming into force at the end of next January,
the new ILAC Arrangement was signed by representatives of accreditation
agencies in twenty-eight countries and means that the results of
tests performed by an accredited laboratory in one country will
be accepted by the signatories of all the other countries. However,
it doesn't necessarily mean that the product's end-users are bound
to accept a foreign calibration certificate, for example.
Former recognition agreements only existed
between some individual agencies or multilaterally through regional
bodies such as APLAC, NACLA and EA. These organizations will continue
to play a crucial role in maintaining confidence in the accreditation
agencies from their region.
Representatives from the following economies
made the commitment:
Australia, Belgium, Brazil, Canada, People's Republic of China,
Czech Republic, Denmark, Finland, France, Germany, Hong Kong China,
India, Ireland, Italy, Japan, Republic of Korea, Netherlands, New
Zealand, Norway, Singapore, South Africa, Spain, Sweden, Switzerland,
Chinese Taipei, United Kingdom, United States of America, Vietnam.
So
what's "Accreditation" anyway?
Accredited laboratories are independently assessed to meet established
international criteria concerning their quality system and technical
competence in the particular field of measurement.
No Recount Needed in ISO9000 Vote !
29 November 2000
In marked contrast to the
shenanigans that afflicted this month's US presidential election,
ISO has announced a landslide victory that has paved the way for
publication of ISO9000:2000 by mid-December.
Of the 63 member bodies that participated in the
two-month ballot, 94% voted to approve the new ISO9000 standard.
However, an undisclosed country was seriously out-of-step with international
opinion concerning the new 9001 and 9004 documents as an amazing
62 positive votes were cast. More surprising still is the revelation
that the single negative vote came from a member who had actually
participated in the standard's development !
Despite the overwhelming result, the versions
to be published next month will be slightly different to the Final
Draft International Standards (FDIS). Minor editorial modifications
have been made to improve clarity of the texts that will be published
as ISO9001:2000 and ISO9004:2000. In the case of ISO9000:2000, more
substantial modifications have been made to the definitions of auditing
terms which is due to the continuing evolution of the other standard
in the ISO 9000 "core series", ISO 19011, Guidelines on Quality
and/or Environmental Management Systems Auditing.
|