Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

This is about par for course for Myki.

https://en.wikipedia.org/wiki/Myki

TL;DR: Melbourne spent over $1.5B developing their very own tap to pay public transport card from scratch, which was and remains slow, glitchy and is already outdated (eg doesn't support payment with contactless credit/debit cards). Meanwhile, Sydney deployed an off the shelf system from Cubic at a fraction of the cost and the rollout was about as smooth as it gets.



Sydney's rollout was definitely not smooth! NSW spent a decade (late 90s - late 00s) trying to launch the "Tcard", after which time they had absolutely nothing to show except wasted $$ millions. Unlike Melbourne, Sydney still had ancient paper tickets well into the 2010s, which weren't even consistent across modes of transport (destination specific "one way / return from x to y" single use tickets for trains, zoned "travel ten" tickets for buses, etc), and which often had to be paid for with exact change in coins (and which often couldn't be purchased in advance). Opal only soft launched in 2012, it wasn't actually available on all modes of transport, for all ticket types, etc, until end of 2014.


Not to mention the massive cost increase to those who previously bought yearly Zone 3 passes. I saw my transit costs increase overnight from $2200 per year to over $3500 with the introduction of Opal.


I don't live in either city, I have family who live in both Melbourne and Sydney so I travel to each city relatively frequently at least a couple of times a year. My experience was Melbourne had their system years before Sydney had Opal cards.

I can distinctly remember first time using tap on/off card to get around Melbourne and thinking how much more convenient it was then needing to use Paper ticket machines Sydney had at the time.

Wikipedia lists 2008 adoption for Melb vs 2012 for Syd that seems to align with my memory. So might have just been curse of them being an early adopter.


You're right on timing, but it wasn't that. The entire rollout was a clusterfuck. There are heaps of news articles from 2006-2008 about it.


Other countries had it before Melbourne and rolled it out smoothly. Hong Kong launched theirs in 1997. 11 years prior. Melbourne could have easily bought it off the shelf. Melbourne was not early by any measure


The issue with myki wasn't the card technology, it was the fact they wanted to have standardised ticketing across the entire state.

eg Tap on a bus, tram and train in Melbourne, get off in Wangaratta and tap on to a bus there.

There was going to be something like 29 zones, and all the requirements / edge cases / mucking around sent the cost through the roof.


What I'm trying to say is that it was mismanagement, which was well known at the time. Hong Kong also has Metro, trains (at the time), buses, trams, ferries. All from different companies. In Melbourne they are actually mostly under the same government agency.


The Hong Kong system's backend, and large parts of the frontend, was also written by an Australian company, ERG at the time, now VIX. So there was plenty of local skills that _could_ have been used.


That's interesting history! It goes to show what could have been if they asked around.


people in hong kong had cellphones before the year 2000


Brisbane had trials for a Cubic system in 2006 with a full rollout by 2008. It went, broadly speaking, OK. The machines at stations were a near unusable mess and the locations you could buy the cards from were ridiculously limited.


It's hilarious that you think the NSW Opal system is better than Myki, because compared to overseas rollouts it's still a trash fire.

There's still no travel card integration with phones, the only mobile device tap that's supported are standard credit cards -- but you can't give one of those to a child because you can't restrict a credit card to only allowing travel! It also doesn't track usage properly, so you can't tell what state you're in (tapped on or off).

They actually did develop a digital wallet card and matching phone app, rolled it out... to some country town... and then quietly dropped it.

My completely unfounded accusation is that someone in Transport NSW was getting kickbacks from Mastercard and/or Visa and a travel card app would have stopped the pork.

The Opal card project cost hundreds of millions of dollars and took way too long to deploy, so calling it a "success" is only in relation to even worse projects.


> There's still no travel card integration with phones

This is not 100% accurate. There is no card integration, but the Opal app does allow you too see pre-paid balances. It does usually show the tapped on/off status, but it may be delayed by a few hours (so it's not completely useful) especially outside of metro areas.

> They actually did develop a digital wallet card

Which never worked on iOS, it was Android only iirc.

> the only mobile device tap that's supported are standard credit cards

You can use Samsung Pay. Make the card inaccessible for normal NFC/touch payments, and only allow Samsung Pay Transit. On iOS I think you can use Express Mode? https://support.apple.com/en-gb/105079

We attempted smart card transport in NSW three times. TCard. There was another one that went bankrupt iirc.

Arguably there is much greater success with Opal vs. Myki.


> because compared to overseas rollouts it's still a trash fire

That's unfair because you are implicitly only comparing against successful rollouts.

Having used a variety of bus systems in South America and Auckland, I can say there are plenty of shitty systems out there.

I tried to use Auckland the other day and couldn't because I didn't have the required bus card. I couldn't use contactless payment from my phone and I had just arrived from overseas so no cash. I was told they don't accept cash anyway. I had to Uber instead.

After dealing with crappy overseas it was eye-opening to have multiple similar problems in my own country (not just bus access) e.g. cash machines don't support contactless so I couldn't get cash - I had lost my CC)


"It also doesn't track usage properly, so you can't tell what state you're in (tapped on or off)."

Right, why can't I use a NFC app (of which I've tried many) to read my Opal card's state - the balance and whether I've tapped on and off? Frankly, it's a damn nuisance.

NFC apps read lots of information about the card, type manufacturer number serial number etc. but nothing that's useful like one's remaining balance. Yes, they can encrypt parts of the card to protect security but there's no need to inconvenience users by not telling them their balance.

Opal is, in effect, a machine-readable only system that takes power away from users. Thus, clearly its ergonomics are unacceptable.

I've also an Opal card with a crack in it which means its internal antenna wire is broken and the card won't respond to the machine. Unfortunately, it has a credit of between $60 and $70 on it and getting that recredited has been such a pain to the extent that I'll have to dissolve the plastic and reconnect the wire. Yes, the card could have been registered but why should I have to tell the government everywhere I go?

Similarly, the Opal system discriminates against those with confession cards in that they're tracked by default and their cards have a louder beep on machines that informs everyone nearby that the person is traveling at a cheaper rate. In my opinion, that's unacceptable and discriminatory (if necessary a ticket inspector could determine that with his handheld reader).

When I was living in Vienna several decades ago there were multiple ways to buy tickets, singles, weekly, monthly and so on. One was a card strip with multiple segments on it. You'd fold over to a blank segment (a split-second operation) and insert it in a time/route stamping machine on the bus/tram or train and that stamp was valid for an hour on any transport even if you changed mid route.

One could see with just one's eyes how much credit one had left and a ticket inspector could glance at the ticket and tell in an instant whether the ticket was valid or not. Also, the route number would tell him if you'd changed your mode of transport or not (that could be useful if one's latter transport was late, which it never was—Vienna's transport system is wonderful).

The ergonomics of that old-fashioned mechanical system were excellent and it worked like a dream. Much of this modern IT tech is done just for the sake of it and for the smart to make a financial killing from the gullible. And I say that as an IT professional.

It really annoys me that more people don't complain about poorly designed IT infrastructure that's barely fit for purpose. Like frogs in heating water, unfortunately we can be cajoled into accepting anything with minimal effort.


The opal app will do what you want. It isn't exactly what you wanted, but it is certainly possible to do


How that if you don't have access to the internet or don't want to link it to an account?

From my experience the card by itself can't provide that information. With a physical time stamp on a card it's immediately clear.


A city should be capable of rolling its own competently at less cost than it takes to engage Cubic to do so.

Cubic is setting things up real nice to have exact data about a lot of peoples travels in cities all around the world. That could be some pretty valuable data.


Why should a city be able to do something more efficiently/effectively than a company which exists for that purpose?


They could seed a company that exists solely for the purpose and wind it up at the completion of the project.

Nevertheless a city is just a company that exists for a purpose.


Sure, except a city is a company that has never done that thing before (vs a company that has done it many times) and a city lacks external motivators like "competition" which drive increased efficiency / reduced cost.


> city lacks external motivators like "competition" which drive increased efficiency / reduced cost.

Which is negated by the city not trying to make a profit and marking up the cost.


They could draw from a lot of people already employed, and they have no incentive to mark up the price.


> They could draw from a lot of people already employed

which means the city would need to pay above the existing wage for those people, since they're _already_ employed and would have no incentive to jump.

> no incentive to mark up the price

and have no incentive to try to save money or find novel efficiencies, since the money spent is not "theirs".


> which means the city would need to pay above the existing wage for those people, since they're _already_ employed and would have no incentive to jump.

What? I'm talking about salaried employees already working for the city.

> and have no incentive to try to save money or find novel efficiencies, since the money spent is not "theirs".

Even if they are not trying to actively save money, they are not trying to actively make a profit so it would still be cheaper.


That's... not how it works.

You can't just throw people at a problem and assume they will get it done for some fixed cost. Experience and existing infrastructure and solutions actually matter, and can make something significantly cheaper than building from scratch.

Let's say you want to bake a loaf of bread. You're not a baker, so you don't have any of the skills, ingredients, or tools needed to bake this loaf. Also, you only want to bake the one loaf (just like building out a contactless ticket system is a single project that shouldn't need to be done over and over again).

Which would you expect to be cheaper and faster: 1. buying everything you need to bake a loaf of bread and doing it yourself or 2. walking to the nearest bakery and buying a loaf there.

Keep in mind that the bakery needs to make a profit, and you do not – so there is no way the bakery will end up being cheaper, right? And surely the seasoned professionals (who already have the necessary tools and ingredients) won't be faster at baking the loaf than you, a complete novice without those things.


> That's... not how it works.

If often is.

Organizations with preexisting staff and knowledge frequently implement things instead of outsourcing.

> Experience and existing infrastructure and solutions actually matter, and can make something significantly cheaper than building from scratch.

Coming up with a secure fare payment and tracking system isn't some cutting edge concept. Work would still be outsourced where necessary, like ordering or manufacturing the hardware.

> You're not a baker, so you don't have any of the skills, ingredients, or tools needed to bake this loaf.

That's where the analogy fails. Municipal governments would have people with the skills, acquiring ingredients and tools could be outsourced as needed, but 'baking a loaf of bread' could still be mostly an internal project.


but the point is why does the municiple gov't have someone with baking skills on staff, when they don't bake bread at all normally?


Because they have bake muffins, pizzas, cakes, pancakes, pasta and pies regularly.

Turns out when you have experience in the general area of baking bread, learning to bake bread isn't that hard.


Your assumptions about the flexibility and transferability of a municipal employee's skillset are doing a lot of heavy lifting.


Not at all. Assuming a city has a mix of people with database, web and security experience is perfectly reasonable. You can definitely find someone capable enough to run manage the project, and as I said what needs to can still be outsourced.


Are these proficient people not already doing something with their time?

If yes, do you hire someone else to replace the work they were doing?

If no, sounds like a lot of slippage.


Have you never worked at a large company?

People don't always have their billable hours fully booked.


Why would a city that has never built a contactless ticket gate system have experience building a contactless ticket gate system?


It's not that hard is the basic answer. They have plenty of people with IT experience capable of managing the project to develop and in house implementation. Even outsourcing what they need to will be cheaper than going with what a company like Cubic offers, where the price is significantly marked up over the actual cost of developing.

It's a better use of tax payer money with the bonus that the city is not dependent on a third party for-profit entity.


Difficulty has almost nothing to do with it. It's about volume. Baking a loaf of bread is not particularly difficult either. But unless you plan to bake many loaves of bread, it is almost certainly neither cheaper nor faster to do so yourself, even if you have experience baking bread. Bakeries also have profit margins, and that changes nothing. All companies make a profit. That doesn't mean they are fleecing you.

Do you make your own clothes? Certainly it must be cheaper to do it yourself rather than buy a shirt from a company, right? Because they have their profits to worry about. Does every city have its own manhole cover factory? Why not? Certainly it must be cheaper to build your own manhole cover factory (forging manhole covers is not a hard problem, after all) rather than pay a company to do it. After all, the manhole cover company is burdened by profit motive, and the city is not.

Repeat ad nauseam.


> it is almost certainly neither cheaper nor faster to do so yourself, even if you have experience baking bread.

You are severely underestimating how much a company will mark up the cost to make a profit.

> Do you make your own clothes? Certainly it must be cheaper to do it yourself rather than buy a shirt from a company, right?

Poor analogy. The fare system being talked about is a big investment that doesn't need to be replaced every year. A closer equivalent would be building your own computer, and yes many people do. The parts they manufacturer they still buy or have made.

> Repeat ad nauseam.

Except your analogy doesn't apply across the board and you know this. If we extrapolate from your reasoning a city shouldn't do anything themselves.


> If we extrapolate from your reasoning a city shouldn't do anything themselves.

exactly correct. A city should be a bureaucratic, administrative organization that collects taxes, and funds projects. These projects should be done by professional project doers - such as construction companies, or IT companies, or bakers. There ought to be many project doers, all competing for the city's funding. The city's job is to find the lowest cost, highest value project doer to give them the project.

The people of the city chooses what projects to fund, and the city takes care of the administrative work of allocating the tax dollars. Nowhere in this loop should a city be keeping employees on hand in the expectation of projects that need doing.


One year my city put the sidewalk snow-clearing out on contract to a private company, but then the sidewalks didn't get cleared between Christmas and New Year, because the company was on holiday during that period.

Someone wrote an editorial that would not pass HN's comment policy, basically to the effect of "who would ever have guessed that it might snow in between Christmas and New Year?" and sidewalk-clearing went back to being a public task.


In my experience public workers are more likely to have public holidays off than private workers.


> exactly correct.

No it isn't. What nonsense. Cities have been doing things themselves all around the world for decades because it generally makes more sense.

> Nowhere in this loop should a city be keeping employees on hand in the expectation of projects that need doing.

They already do that, look at just IT for example, they maintain websites, databases, etc. They don't outsource everything, that's ridiculous.


Metcards were more reliable than either.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: