Ventra - Bugs, Feedback, and Questions

[busfan2847]

Another Ventra delay. Metra is putting off for a month the role out of disposable Ventra cards for Link-Up and Plus Bus. It was supposed to start with December tickets but will now start with January.

"The transition for Link-Up and PlusBus cards from magnetic stripe cards to the new Ventra-branded cards has been delayed by one month. That means Metra Monthly Pass users who buy a Link-Up or PlusBus card for connecting travel to the CTA and Pace will receive a magnetic stripe card in December and the new Ventra-compatible cards starting in January.

The December PlusBus magnetic stripe card will look the same. However, the December Link-Up magnetic stripe card will look different: it will say “30-day pass” on the front and back, but printed on the back will be “Metra Link-Up Valid Dec. 2013.” It will be valid for the entire month. These are the cards supplied to Metra by the CTA."

[Amtrak41]

Putting off "for a month" is a politically correct way of rejecting it without saying so explicitly, and give the illusion (like ACA) that everything will be fine by then. It will just get put off month by month.

[Busjack]

Putting off "for a month" is a politically correct way of rejecting it without saying so explicitly, and give the illusion (like ACA) that everything will be fine by then. It will just get put off month by month.

Some of that, but also that coordination between the transit agencies (here all 3) is poor. Like the Transit Benefits plan, it doesn't look like CTA really took those two interagency fare products into account.

The other thing not said is that the two joint products can't be put on a single Ventra card.

[jajuan]

Putting off "for a month" is a politically correct way of rejecting it without saying so explicitly, and give the illusion (like ACA) that everything will be fine by then. It will just get put off month by month.

Again for the umpteenth time the comparison is apples and oranges. The federal government acknowledged there were problems right away and set forth to fix those problems almost immediately. CTA did not do the same in the case of Ventra and what should have only been minor glitches ballooned into major crises. And few of those crises such as the mailing of Ventra cards for a good number riders still has not been fixed. ACA does have measurable improvements and accomplishments in place that the media is not reporting enough about in favor of the ridiculous tendency to keep looking for controversy to outdo one or more news rivals for ratings. You got the fact that young people up to 26 years old can remain on a parent's insurance; children with pre-existing conditions can't be denied coverage, the lifetime coverage caps are gone; the same should be true for adults with policies effective in January: insurance companies have to spend more of the money you pay to them on your actual healthcare; you can't be dropped indiscriminately if you get sick and they have to meet minimum standards of coverage which is the true reason folks were getting those cancellation notices that caused all the recent raucus. And if there were no real improvements in ACA new enrollments for insurance halfway through this month would not already be double of that reported for all of October. With the case of CTA and the Ventra rollout, most of the biggest reported problems are still there with Chicago local NBC station reporting that a bunch of cards out in circulation are working despite having humongous negative balances, the largest one reported being -$272.50. I think there are enough differences in the handling of problems in both respective rollouts to put to the rest the notion that both cases are anywhere similar.

[Busjack]

Again for the umpteenth time the comparison is apples and oranges. .... With the case of CTA and the Ventra rollout, most of the biggest reported problems are still there with Chicago local NBC station reporting that a bunch of cards out in circulation are working despite having humongous negative balances, the largest one reported being -$272.50. I think there are enough differences in the handling of problems in both respective rollouts to put to the rest the notion that both cases are anywhere similar.

I don't want to get into ACA, but the apples and oranges part here is that "what is being put off for a month" is issuing these passes in RFID form instead of mag stripe card. Does not indicate any computer problem, but a problem in getting the new cards, since a pass in RFID form would presumably be accepted.

Apparently, though, delaying pulling the plug on the mag stripe card vending machines and readers enabled CTA to continue using the mag stripe cards to paper over defects in the Transit Benefits program and these passes.

On the last point, the report I saw had Tammy Chase calling Robert Kelly a liar, but that's quite a stretch for someone whose veracity has come under question. A similar incident was hashed out on the CTA Tattler a month ago, but I assume that since part of the labor agreement was that Customer Assistants would be part of the bargaining unit, they are telling Kelly what they observe. The only question is whether the customer or Cubic will be responsible for the big negative balances, but they certainly should be auditable. It seems like Kelly was just reinforcing what the RTA said.

[Vinny]

CTA has released a status report on Ventra and stuff that needs to be fixed:

http://www.transitchicago.com/assets/1/media_relations_documents/Weekly_Performance_NEW_VERSION_112213_FINAL.pdf

Looks like they have rolled out software updates to the card readers at rail stations to bring the transaction processing time down under 1 second in the vast majority of cases (95%). The same update is in the process of being rolled out to buses, but so far, only about 56% of transactions are processed in less than a second (probably because of the cell communication made by bus-mounted readers).

There's some stuff about Ventra accounts and call hold times too.

[owine]

CTA has released a status report on Ventra and stuff that needs to be fixed:

http://www.transitchicago.com/assets/1/media_relations_documents/Weekly_Performance_NEW_VERSION_112213_FINAL.pdf

Looks like they have rolled out software updates to the card readers at rail stations to bring the transaction processing time down under 1 second in the vast majority of cases (95%). The same update is in the process of being rolled out to buses, but so far, only about 56% of transactions are processed in less than a second (probably because of the cell communication made by bus-mounted readers).

There's some stuff about Ventra accounts and call hold times too.

I certainly noticed Ventra readers working much better at rail stations in the past week. 95% of rail taps are processed within a second according to that report.

[Busjack]

CTA has released a status report on Ventra and stuff that needs to be fixed:

http://www.transitchicago.com/assets/1/media_relations_documents/Weekly_Performance_NEW_VERSION_112213_FINAL.pdf

...

There's some stuff about Ventra accounts and call hold times too.

The issue with the hold times is whether the 5 minute standard for Cubic being paid is based on this, which is a 24 hour average, or the admitted but not reported "average hold times are higher during the daytime," especially given the report that the call center personnel is not there 24/7. Obviously, a call made after hours shunted to voice mail is going to drop the average significantly. They should have reported the peak hold time, especially on the days when the call volume in the preceding graph exceeded the average. From that graph, it looks like customers figured out not to call on weekends.

The other notable statement is that 100% of the Ventra cards to CCP and registered CC holders have been shipped. Apparently, if you do not have it by now, don't expect it in the mail, and it says to go out and purchase one. So much for the waiver of the $5 deposit.

[Busjack]

Tribune article with all the contract metrics, if anyone can understand that. Note that Hilkevitch make the point that it is Cubic's numbers, not CTA's numbers in the presentation, and the point I made earlier about average phone hold as opposed to peak.

Also, I don't know about Cubic saying it is on time, when the preexisting schedule would have cut off loading preexisting media by now.

[sw4400]

Getting on the bus hasn't been a problem with me... but this is a head-scratcher, for sure...

I boarded a #9 Ashland bus this afternoon about 12p at Irving Park. When I checked my account, this is what I saw...

Ventra 11-24-13.bmp

[Amtrak41]

Would Ventra be in agreement with the location of the bus as stated by Bus Tracker ?

[Busjack]

Would Ventra be in agreement with the location of the bus as stated by Bus Tracker ?

It was earlier surmised that it had to be connected to the GPS in some manner so as to show the route number on the account. The question to sw would be whether the destination sign also was fouled up.

Otherwise, the question to the drivers would be whether one has to log into Ventra with the route number, and does it take into account interlines (which it certainly would have to do on Pace).

[sw4400]

It was earlier surmised that it had to be connected to the GPS in some manner so as to show the route number on the account. The question to sw would be whether the destination sign also was fouled up.

Otherwise, the question to the drivers would be whether one has to log into Ventra with the route number, and does it take into account interlines (which it certainly would have to do on Pace).

That's one mystery that'll remain unsolved, I guess. I saw the bus turning from Irving Park onto Ashland to pick up passengers. I knew I only had a very limited time to catch it before it departed, so I just ran to get the bus and didn't bother to look at the destination signs. What I do know is that the announcement system was working on the inside, as it was calling the street names out before we reached them and they were showing up on the display. I don't think that means that the exterior displays are working, but at least Clever Devices knows where the bus is, otherwise there would be no announcements or street names in the displays, just the date and time.

[trainman8119]

I think you should contest the charge...however you were using a 7 day so

what's the difference.

[Busjack]

...., but at least Clever Devices knows where the bus is, otherwise there would be no announcements or street names in the displays, just the date and time.

That's probably what counts, as BusTracker treats on off route bus as nonexistent.

[brownliner]

I think you should contest the charge...however you were using a 7 day so

what's the difference.

What's to contest? He got on a bus. He got charged for the bus ride (well, he would have if we were pay per ride). Does it matter to anyone, except maybe a divorce lawyer, which bus it really was? Chicago card never reported what bus you were on, just 'cta bus' or 'pace bus'. 'L' stops it normally got right.

So the ventra reader doesn't know what the clever device does. That's interesting, but not terribly surprising. (or it knows, but can't deal with interlines. Are there any 9/49 interlines?)

[jajuan]

Getting on the bus hasn't been a problem with me... but this is a head-scratcher, for sure...

I boarded a #9 Ashland bus this afternoon about 12p at Irving Park. When I checked my account, this is what I saw...

It's similar to what I reported happened with me a couple of months ago when a 22 Clark I boarded on a weekday just after 7 PM showed up as a 147 Outer Drive Express and on a separate occasion before that a 135 Clarendon/LaSalle Express I boarded showed up as a 148 Clarendon/Michigan Express on my account. I joked at the time that apparently Ventra didn't know the difference between Clarendon/LaSalle Express and Clarendon/Michigan Express.

[jajuan]

What's to contest? He got on a bus. He got charged for the bus ride (well, he would have if we were pay per ride). Does it matter to anyone, except maybe a divorce lawyer, which bus it really was? Chicago card never reported what bus you were on, just 'cta bus' or 'pace bus'. 'L' stops it normally got right.

So the ventra reader doesn't know what the clever device does. That's interesting, but not terribly surprising. (or it knows, but can't deal with interlines. Are there any 9/49 interlines?)

I had a suspicion before that it might be possible, as there were times I noticed the same buses from 74th that would be on the 49 at night during owl service would be on the 9 the very next morning in about the same order.

[Busjack]

In the foul up du jour, federal government employees discover that Ventra will accept their employee ID cards, even though there isn't a bank account associated with it. The article also says that the employees could get in trouble if they abuse their credentials.

The more confusing part of the article is while it explains that the system has to go to the server to verify balances and make deductions, somehow some of the transaction is locally processed on the reader.

[AppleBee]

I am sure that Cubic would never do this, but it crosses my mind that if
someone told you that you would start getting half a billion dollars in
payments if you got your processing time down to under 2.5 seconds, you
might think of just setting a timer that automatically flashes the "go"
light at 2.4 seconds whether you are done processing or not. It might
cost you a few thousand dollars in reimbursements, but what's that
compared to half a billion? And customers might get confused by
negative balances, but you could spin that as a customer service
feature. And after the payments start, you might adjust your system to
operate more strictly. But, again to be clear, this is a fantasy I just
thought up. It is not something that anyone is doing or anyone would
really do.

[Busjack]

I am sure that Cubic would never do this, but it crosses my mind that if

someone told you that you would start getting half a billion dollars in

payments if you got your processing time down to under 2.5 seconds, you

might think of just setting a timer that automatically flashes the "go"

light at 2.4 seconds whether you are done processing or not. It might

cost you a few thousand dollars in reimbursements, but what's that

compared to half a billion? ...

The half billion is in monthly installments over 10 or 12 years. So, how many false green lights are there going to be over 10 or 12 years? Remember the article about that one outage at 60 rail stations resulted in an estimated 15,000 free rides.

The real question is whether CTA can audit the missed fares. One story said something about using cameras, but it should be easier just to use the gps counters in the bus door wells and any counter in the turnstiles.

The real question I posed before was the part about locally processed transactions later being flagged.Like I said, the article didn't say how that worked, and whether that is, in effect, what you said.

[AppleBee]

But, according to Jon Hilkevitch, once the CTA declares that the goals are met "the transit system is obligated to begin paying Cubic at least $2.5 million a month, a bottom line figure that could be hard to reduce regardless of future problems."

A fictional hypothetical company would just need to keep up the green lights until the CTA declares the goals met and the money would then start rolling in for 10 to 12 years. They wouldn't have to keep the green lights on for 10 to 12 years.

[sw4400]

Rather than starting a new topic with a poll, I want to get the forums' opinion of the Ventra Card system... my house is pretty much divided...

• Get rid of Ventra and go back to Magnetic Stripe Cards/Chicago Cards? 0 Votes
• Give Ventra a chance to be ironed out. The bugs will be fixed eventually: 1 Vote
• Look for a new vendor to create a tap and go system to work with: 0 Votes

In all honesty, we're going to a tap and go system one way or the other. I remember when Chicago Card and Chicago Card Plus came out, there were lots of problems with them. The CTA got them ironed out and it worked flawlessly. I think the days of Magnetic Stripe Cards is done. Ventra does have it's kinks, but like Chicago Card and Chicago Card Plus in it's infancy, it'll all iron out in due time.

[Busjack]

Rather than starting a new topic with a poll, I want to get the forums' opinion of the Ventra Card system... my house is pretty much divided...

• Get rid of Ventra and go back to Magnetic Stripe Cards/Chicago Cards? 0 Votes
• Give Ventra a chance to be ironed out. The bugs will be fixed eventually: 1 Vote
• Look for a new vendor to create a tap and go system to work with: 0 Votes

In all honesty, we're going to a tap and go system one way or the other. I remember when Chicago Card and Chicago Card Plus came out, there were lots of problems with them. The CTA got them ironed out and it worked flawlessly. I think the days of Magnetic Stripe Cards is done. Ventra does have it's kinks, but like Chicago Card and Chicago Card Plus in it's infancy, it'll all iron out in due time.

I don't think it is going back to the old media, so the choice (as I indicated before) is either Cubic fixes it, or CTA tells Cubic to get its equipment out of here. Not much different than Emanuel dumping the red light camera company, although his motivations were different.

But your poll doesn't tale into account all the consumer fraud associated with Ventra, both on Cubic's and CTA's account. Cubic's might have been accidental, CTA's was not.

[brownliner]

 

In the foul up du jour, federal government employees discover that Ventra will accept their employee ID cards, even though there isn't a bank account associated with it. The article also says that the employees could get in trouble if they abuse their credentials.

The more confusing part of the article is while it explains that the system has to go to the server to verify balances and make deductions, somehow some of the transaction is locally processed on the reader.

 

There is a provision in mastercard's RFID payment system (I'm too lazy to dig my ventra card out to figure out what the heck they call it.) to allow for offline transactions. In an offline transaction, there's no round trip between the reader and the backend network. It's all done locally, and should take less than second. The chip on the card has a set of parameters that specify when it's allowed to do offline transactions. The card issuer gets to set them, and can do different things based on the account the card is associated with.

It's possible to set whether the card can do offline at all, the maximum number of consecutive offline transactions before an online transaction is forced, the maximum spend per offline transaction, and the maximum spend (total, across transactions) before an on line transaction is forced. The reader uses that, and its local information (it can have a 'hotlist' of known bad cards, for instance) to decide whether to approve, decline, or force online for approval/denial. If an online transaction is forced, you have to wait for that, and the speed of that depends on lots of things, including the speed of the data link.

It's pretty clear that Ventra is designed with expectation of doing most transactions offline. Fixed readers, like at train stations might do online, to reset the counter so more bus transactions can be online.

It's not impossible Cubic is playing games with that behavior right now. It's not impossible they fixed a bug, that was causing offline eligible transactions to be done online.