better than average article on why the federal payroll system is a mess
There have been a bunch of headlines about the new federal payroll system "Phoenix" is screwing up a lot. This article does a reasonable job of describing how the problems happened. <http://www.itworldcanada.com/blog/phoenix-payroll-report-by-michael-wernick-the-clerk-of-the-privy-council/385370>
On 08/12/2016 11:13 AM, D. Hugh Redelmeier via talk wrote:
There have been a bunch of headlines about the new federal payroll system "Phoenix" is screwing up a lot. This article does a reasonable job of describing how the problems happened.
<http://www.itworldcanada.com/blog/phoenix-payroll-report-by-michael-wernick-the-clerk-of-the-privy-council/385370> --- Talk Mailing List talk@gtalug.org https://gtalug.org/mailman/listinfo/talk Interesting.
I have a slightly different take. Through my wife's work I have had a chance to see lots of government projects (mostly Ontario) and I have come to the conclusion that the insistance of project managers to go with safe products like Oracle and Microsoft and Cisco are the reason lots of projects are off the rails. Big companies know how to bill and charge for changes but the products and services seldom come close to the sales pitch. But nobody will get fired for buying IBM/Cisco/Oracle/Microsoft..... -- Alvin Starr || voice: (905)513-7688 Netvel Inc. || Cell: (416)806-0133 alvin@netvel.net ||
On Fri, Aug 12, 2016 at 10:41 AM, Alvin Starr via talk <talk@gtalug.org> wrote:
On 08/12/2016 11:13 AM, D. Hugh Redelmeier via talk wrote:
There have been a bunch of headlines about the new federal payroll system "Phoenix" is screwing up a lot. This article does a reasonable job of describing how the problems happened.
<http://www.itworldcanada.com/blog/phoenix-payroll-report-by-michael-wernick-the-clerk-of-the-privy-council/385370> --- Talk Mailing List talk@gtalug.org https://gtalug.org/mailman/listinfo/talk
Interesting.
I have a slightly different take.
Through my wife's work I have had a chance to see lots of government projects (mostly Ontario) and I have come to the conclusion that the insistance of project managers to go with safe products like Oracle and Microsoft and Cisco are the reason lots of projects are off the rails. Big companies know how to bill and charge for changes but the products and services seldom come close to the sales pitch.
But nobody will get fired for buying IBM/Cisco/Oracle/Microsoft.....
My suggestion is that management get paid on effectiveness - - ie poor outcomes first wage reductions (starting at 25% and growing quickly) and termination for outcomes that just don't work (that's with an independent review to make sure that something wasn't torpedoed because of an even goofier more senior exec!). How would that work you think? Dee
On 12 August 2016 at 11:49, o1bigtenor via talk <talk@gtalug.org> wrote:
On Fri, Aug 12, 2016 at 10:41 AM, Alvin Starr via talk <talk@gtalug.org> wrote:
On 08/12/2016 11:13 AM, D. Hugh Redelmeier via talk wrote:
There have been a bunch of headlines about the new federal payroll system "Phoenix" is screwing up a lot. This article does a reasonable job of describing how the problems happened.
<http://www.itworldcanada.com/blog/phoenix-payroll-report-by-michael-wernick-the-clerk-of-the-privy-council/385370> Interesting.
I have a slightly different take.
Through my wife's work I have had a chance to see lots of government projects (mostly Ontario) and I have come to the conclusion that the insistance of project managers to go with safe products like Oracle and Microsoft and Cisco are the reason lots of projects are off the rails. Big companies know how to bill and charge for changes but the products and services seldom come close to the sales pitch.
But nobody will get fired for buying IBM/Cisco/Oracle/Microsoft.....
My suggestion is that management get paid on effectiveness - - ie poor outcomes first wage reductions (starting at 25% and growing quickly) and termination for outcomes that just don't work (that's with an independent review to make sure that something wasn't torpedoed because of an even goofier more senior exec!).
How would that work you think?
That leads to cancelling the project throughly and regularly, and continuing to run the old software until it falls prey of some equivalent to the 2038 problem, and they have nobody that comprehends anything deep about the old system because those people all retired in the 1990's. Ordering a government bureaucracy to not behave the way a government bureaucracy behaves is as foolish as trying to feed cats a vegetarian diet. Cats are carnivores, and we know to call those that have the delusion that it is a good idea to make them into vegetarians delusional fools. It seems to me that the government had few real options in the matter; the Payroll Problem is big enough that it properly requires the business help of an organization like IBM. (Other plausible options would include CGI, Accenture, PWC, Deloitte Consulting, KPMG, but those don't have the selections of hardware and software that IBM would offer, so I'm completely unsurprised at IBM falling to the top of the list.) They can't contract it all out to Ceridian/ADP, which is what a whole lot of business do, because there's a large enough set of people doing Actually Secret Stuff, between payroll of: - Prison staff - Diplomatic staff - RCMP staff - CSIS staff (honest-to-goodness spies among them) - Betcha paid informants need to get, um, paid... Actually, my straight response to "be more business like" is pretty much pointing at this set of secret squirrels... Businesses don't have these bits, and are not allowed to, pretty much. Government isn't a business, which is a good reason NOT to put a "businessman" in charge. -- When confronted by a difficult problem, solve it by reducing it to the question, "How would the Lone Ranger handle this?"
Chris, I think you're on to something here. Its the execs who are the problem. The fundamental problem is that the current crop of executives have no understanding of technology and yet technology is a key driver of productivity. These people are well versed in accounting principles and know who to befriend in order to raise their executive bonuses but little else. /gary On 16-08-12 12:34 PM, Christopher Browne via talk wrote:
On 12 August 2016 at 11:49, o1bigtenor via talk <talk@gtalug.org> wrote:
On 08/12/2016 11:13 AM, D. Hugh Redelmeier via talk wrote:
There have been a bunch of headlines about the new federal payroll system "Phoenix" is screwing up a lot. This article does a reasonable job of describing how the problems happened.
<http://www.itworldcanada.com/blog/phoenix-payroll-report-by-michael-wernick-the-clerk-of-the-privy-council/385370> Interesting.
I have a slightly different take.
Through my wife's work I have had a chance to see lots of government projects (mostly Ontario) and I have come to the conclusion that the insistance of project managers to go with safe products like Oracle and Microsoft and Cisco are the reason lots of projects are off the rails. Big companies know how to bill and charge for changes but the products and services seldom come close to the sales pitch.
But nobody will get fired for buying IBM/Cisco/Oracle/Microsoft..... My suggestion is that management get paid on effectiveness - - ie poor outcomes first wage reductions (starting at 25% and growing quickly) and termination for outcomes that just don't work (that's with an independent review to make sure
On Fri, Aug 12, 2016 at 10:41 AM, Alvin Starr via talk <talk@gtalug.org> wrote: that something wasn't torpedoed because of an even goofier more senior exec!).
How would that work you think? That leads to cancelling the project throughly and regularly, and continuing to run the old software until it falls prey of some equivalent to the 2038 problem, and they have nobody that comprehends anything deep about the old system because those people all retired in the 1990's.
Ordering a government bureaucracy to not behave the way a government bureaucracy behaves is as foolish as trying to feed cats a vegetarian diet. Cats are carnivores, and we know to call those that have the delusion that it is a good idea to make them into vegetarians delusional fools.
It seems to me that the government had few real options in the matter; the Payroll Problem is big enough that it properly requires the business help of an organization like IBM. (Other plausible options would include CGI, Accenture, PWC, Deloitte Consulting, KPMG, but those don't have the selections of hardware and software that IBM would offer, so I'm completely unsurprised at IBM falling to the top of the list.)
They can't contract it all out to Ceridian/ADP, which is what a whole lot of business do, because there's a large enough set of people doing Actually Secret Stuff, between payroll of: - Prison staff - Diplomatic staff - RCMP staff - CSIS staff (honest-to-goodness spies among them) - Betcha paid informants need to get, um, paid...
Actually, my straight response to "be more business like" is pretty much pointing at this set of secret squirrels... Businesses don't have these bits, and are not allowed to, pretty much. Government isn't a business, which is a good reason NOT to put a "businessman" in charge.
On 08/12/2016 12:34 PM, Christopher Browne via talk wrote:
On 12 August 2016 at 11:49, o1bigtenor via talk <talk@gtalug.org> wrote:
On 08/12/2016 11:13 AM, D. Hugh Redelmeier via talk wrote:
There have been a bunch of headlines about the new federal payroll system "Phoenix" is screwing up a lot. This article does a reasonable job of describing how the problems happened.
<http://www.itworldcanada.com/blog/phoenix-payroll-report-by-michael-wernick-the-clerk-of-the-privy-council/385370> Interesting.
I have a slightly different take.
Through my wife's work I have had a chance to see lots of government projects (mostly Ontario) and I have come to the conclusion that the insistance of project managers to go with safe products like Oracle and Microsoft and Cisco are the reason lots of projects are off the rails. Big companies know how to bill and charge for changes but the products and services seldom come close to the sales pitch.
But nobody will get fired for buying IBM/Cisco/Oracle/Microsoft..... My suggestion is that management get paid on effectiveness - - ie poor outcomes first wage reductions (starting at 25% and growing quickly) and termination for outcomes that just don't work (that's with an independent review to make sure
On Fri, Aug 12, 2016 at 10:41 AM, Alvin Starr via talk <talk@gtalug.org> wrote: that something wasn't torpedoed because of an even goofier more senior exec!).
How would that work you think? That leads to cancelling the project throughly and regularly, and continuing to run the old software until it falls prey of some equivalent to the 2038 problem, and they have nobody that comprehends anything deep about the old system because those people all retired in the 1990's. So your talking about standard business practice for lots of businesses. I am still supporting software I wrote 25 years ago and can't get it killed.
Ordering a government bureaucracy to not behave the way a government bureaucracy behaves is as foolish as trying to feed cats a vegetarian diet. Cats are carnivores, and we know to call those that have the delusion that it is a good idea to make them into vegetarians delusional fools.
Large organizations have a whole raft of issues relating to their size the problems that governments have is that they are just about the largest of organizations. On the other hand large organizations have the resources to take part in large projects where large financial commitments must be made.
It seems to me that the government had few real options in the matter; the Payroll Problem is big enough that it properly requires the business help of an organization like IBM. (Other plausible options would include CGI, Accenture, PWC, Deloitte Consulting, KPMG, but those don't have the selections of hardware and software that IBM would offer, so I'm completely unsurprised at IBM falling to the top of the list.)
There seems to be a common misconception that commercial products are better than in-house designed products. Governments could easily justify having their own in-house design and development staff but when projects go bad as often happens it becomes public knowledge and some bright guy says "you should have used a commercial product ....". The thing is that you seldom hear about the big screw-ups at large corporations in the way of failed projects. The shareholder reports don't have a category for money lost on bad projects.
They can't contract it all out to Ceridian/ADP, which is what a whole lot of business do, because there's a large enough set of people doing Actually Secret Stuff, between payroll of: - Prison staff - Diplomatic staff - RCMP staff - CSIS staff (honest-to-goodness spies among them) - Betcha paid informants need to get, um, paid...
That's not completely true. Governments contract with companies to undertake lots of tasks that require security. Who builds the spy-satellites? I use to work for a NATO hardware contractor and had access to secret information.
Actually, my straight response to "be more business like" is pretty much pointing at this set of secret squirrels... Businesses don't have these bits, and are not allowed to, pretty much. Government isn't a business, which is a good reason NOT to put a "businessman" in charge.
Governments are a business in many respects. The thing is that the corporate officers(politicians) are not paid in same way that an incorporated business CEO/director is. So the pressures driving decisions are completely different. Some times that difference is a good thing. -- Alvin Starr || voice: (905)513-7688 Netvel Inc. || Cell: (416)806-0133 alvin@netvel.net ||
On 12/08/16 12:34 PM, Christopher Browne via talk wrote:
On 12 August 2016 at 11:49, o1bigtenor via talk <talk@gtalug.org> wrote:
On 08/12/2016 11:13 AM, D. Hugh Redelmeier via talk wrote:
There have been a bunch of headlines about the new federal payroll system "Phoenix" is screwing up a lot. This article does a reasonable job of describing how the problems happened.
<http://www.itworldcanada.com/blog/phoenix-payroll-report-by-michael-wernick-the-clerk-of-the-privy-council/385370> Interesting.
I have a slightly different take.
Through my wife's work I have had a chance to see lots of government projects (mostly Ontario) and I have come to the conclusion that the insistance of project managers to go with safe products like Oracle and Microsoft and Cisco are the reason lots of projects are off the rails. Big companies know how to bill and charge for changes but the products and services seldom come close to the sales pitch.
But nobody will get fired for buying IBM/Cisco/Oracle/Microsoft..... My suggestion is that management get paid on effectiveness - - ie poor outcomes first wage reductions (starting at 25% and growing quickly) and termination for outcomes that just don't work (that's with an independent review to make sure
On Fri, Aug 12, 2016 at 10:41 AM, Alvin Starr via talk <talk@gtalug.org> wrote: that something wasn't torpedoed because of an even goofier more senior exec!).
How would that work you think? That leads to cancelling the project throughly and regularly, and continuing to run the old software until it falls prey of some equivalent to the 2038 problem, and they have nobody that comprehends anything deep about the old system because those people all retired in the 1990's.
Ordering a government bureaucracy to not behave the way a government bureaucracy behaves is as foolish as trying to feed cats a vegetarian diet. Cats are carnivores, and we know to call those that have the delusion that it is a good idea to make them into vegetarians delusional fools.
It seems to me that the government had few real options in the matter; the Payroll Problem is big enough that it properly requires the business help of an organization like IBM. (Other plausible options would include CGI, Accenture, PWC, Deloitte Consulting, KPMG, but those don't have the selections of hardware and software that IBM would offer, so I'm completely unsurprised at IBM falling to the top of the list.)
They can't contract it all out to Ceridian/ADP, which is what a whole lot of business do, because there's a large enough set of people doing Actually Secret Stuff, between payroll of: - Prison staff - Diplomatic staff - RCMP staff - CSIS staff (honest-to-goodness spies among them) - Betcha paid informants need to get, um, paid...
Actually, my straight response to "be more business like" is pretty much pointing at this set of secret squirrels... Businesses don't have these bits, and are not allowed to, pretty much. Government isn't a business, which is a good reason NOT to put a "businessman" in charge.
I have a former boss who now works for federal purchasing, specializing in software. If the old grey heads would listen to the middle-aged females (or even the males, of the era) they would not be looking at career-threatening mistakes. Alas, listening to one's subordinates in an organization with more than 100 members is a career *ending* mistake. --dave -- David Collier-Brown, | Always do right. This will gratify System Programmer and Author | some people and astonish the rest davecb@spamcop.net | -- Mark Twain
On 12 August 2016 at 11:13, D. Hugh Redelmeier via talk <talk@gtalug.org> wrote:
There have been a bunch of headlines about the new federal payroll system "Phoenix" is screwing up a lot. This article does a reasonable job of describing how the problems happened.
I love the bits of recommendations to the PM in wee italicized quips... Particularly the "I recommend you not acknowledge the existence of this problem." The notion of there being varying interpretations of employment contracts in different locations is... interesting... I don't envy them the problems; payroll is a tougher problem than it seems, and transitioning to a new system is a mighty difficult task. Been there, lost some hair to it (I worked on the transitions for American Airlines and Sabre). With 300,000 civil servants spread across the country, and hence numerous legal jurisdictions (10 provinces, 3 territories, plus diplomatic staff residing elsewhere), and some fairly large number of unions and hence labour contracts, there's a LOT of complexity. If they're imagining they'll cut from 2400 payroll staff to 600-ish, then a lot of the resulting rules need to be encoded in software/configuration and validated, which is pretty daunting. Anyone that has in their head that they "just" need to configure several copies of QuickBooks Payroll, I suggest you smack your head against a wall sufficiently vigorously to thoroughly knock such thoughts out... -- When confronted by a difficult problem, solve it by reducing it to the question, "How would the Lone Ranger handle this?"
On Aug 12, 2016, at 11:13, D. Hugh Redelmeier via talk <talk@gtalug.org> wrote: <http://www.itworldcanada.com/blog/phoenix-payroll-report-by-michael-wernick-the-clerk-of-the-privy-council/385370>
Some of my thoughts: 1. I don't like the literary form of this article. Too much theatre. I prefer non-fiction, not info-fiction. Even the URL misleads. 2. (Why) was there no parallel test period? At least for a stratified sample. That might have shown the unanticipated problems (which should have been anticipated). 3. When a sufficiently complex system is replaced, it should not be assumed that the old system was completely or consistently correct. (CF: the allusions to interpretation differences.) Systems that operate for a long time tend to become corrupt. Errors, ambiguities, interpretations, may creep in. Some of these are never discovered. Some are discovered and corrected. Others are discovered and covered up -- the error is propagated forward to avoid having to deal with the consequences. A consistent replacement system for an inconsistent system will act differently, and efforts to retain inconsistencies may result in instabilities and failures. 4. The vendor shares in the responsibility for taking on a task that contained "unanticipated complexities". There should have been vendor-penalties for such failures. And a responsible vendor would do enough due diligence to anticipate "complexities" early. -- Peter
| From: Peter Renzland via talk <talk@gtalug.org> | 1. I don't like the literary form of this article. Too much theatre. | I prefer non-fiction, not info-fiction. Even the URL misleads. Each to his own taste. I thought it was fine but I didn't think of it as a tour de force | 2. (Why) was there no parallel test period? At least for a stratified | sample. That might have shown the unanticipated problems (which | should have been anticipated). The article hinted that there was one. PSAC wanted to delay the second phase of the rollout because thousands of its members had already been shafted on payments from Phoenix. | 3. When a sufficiently complex system is replaced, it should not be | assumed that the old system was completely or consistently correct. Right. But there is an effect like common law. If something goes on for long enough, it seems like the Right Thing. | (CF: the allusions to interpretation differences.) Systems that | operate for a long time tend to become corrupt. Errors, ambiguities, | interpretations, may creep in. Some of these are never discovered. In theory, fixing these things is good. But many people might be agrieved. Just imagine the screems if police started enforcing the actual speed limits? Or charged everyone running red lights. | 4. The vendor shares in the responsibility for taking on a task that | contained "unanticipated complexities". There should have been | vendor-penalties for such failures. And a responsible vendor would | do enough due diligence to anticipate "complexities" early. You worked in government a long time. Not me. So your guesses are more likely correct than mine. My understanding is that RFQs are always overly precise so that the system appears fair: all are bidding on the same thing. The downside is that the customer is on the hook for changes. Changes become a profit centre because there is no competition in this phase. To be fair, capturing the behaviour of the current system and capturing the behaviour desired is a very lare part of the work to be done. It seems unreasonable to do that outside of some contract (perhaps a separate one).
On Aug 12, 2016, at 14:17, D. Hugh Redelmeier via talk <talk@gtalug.org> wrote:
You worked in government a long time. Not me. So your guesses are more likely correct than mine.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 22/08/16 10:29 AM, Peter Renzland via talk wrote:
On Aug 12, 2016, at 14:17, D. Hugh Redelmeier via talk <talk@gtalug.org> wrote:
You worked in government a long time. Not me. So your guesses are more likely correct than mine.
http://dilbert.com/strip/2016-08-16
:-(
https://www.youtube.com/watch?v=gJ1Mz7kGVf0 Nikola Tesla vs Thomas Edison ERB - -- Daniel Villarreal http://www.youcanlinux.org youcanlinux at gmail.com PGP key 2F6E 0DC3 85E2 5EC0 DA03 3F5B F251 8938 A83E 7B49 https://pgp.mit.edu/pks/lookup?op=get&search=0xF2518938A83E7B49 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEzBAEBCAAdBQJXuxNgFhx5b3VjYW5saW51eEBnbWFpbC5jb20ACgkQ8lGJOKg+ e0nJBgf+Oc+XOouH5Xl3EpCadozvzOD0c3uUIZ5bLO/fF9l4zCH2s2J4g2WvpMDj EzkP+fUYOy7zzf+9XkeOBcTRs0uoIm47K8tCNi/O7TXF1wR+b5JlD34ATHKDv0EM kAzlHTRIn7xMDdgwS5x3GmmGQFGhyOkuDwVfxyY4Q0q6Dfc0mEsxOIdq3gqSl0jM Ad0xm13JcgxurLM7sWQYfehycZi4ZG36C7g74Cuqi09rTOx6CDrA5hpX10CNcSzv 2DgvPaVonTtL5hCzZlZObrEHhGmYKG2JO7rP/ffSh4ZhsuND0MsLOOq9whsIQL3q 9FusUFHuFRVMQI3wX4jWunFYXt3DDQ== =3OFj -----END PGP SIGNATURE-----
participants (9)
-
Alvin Starr -
Christopher Browne -
D. Hugh Redelmeier -
Daniel Villarreal -
David Collier-Brown -
Gary -
James Knott -
o1bigtenor -
Peter Renzland