I don't care if 18F makes money, and in some cases it may make sense for them not to.
In a perfect market, they would be able to sell their services at the same price as the value of those services. So if they build a system for the Veterans Administration (VA) that saves $1 million, they should get $1 million for it. Their EOY revenue would reflect their value created.
But they are actually selling to themselves. They are a part of the federal government, and their customers are other agencies. So the price they should charge is actually (value created - effort to get paid).
If they waste $500,000 of man-hours to get that $1m, then the government loses $500k for no reason, other than for 18F to be able to point to $1m in revenue as proof of services provided.
The government as a whole also loses out if price is an issue for the buyer, but a net win for the whole organization. If bureaucracy means that a department isn't willing to spend $1m to reap $2m in savings, because maybe the benefits are nebulous or internal politics or any of the other crazy things that people do in big organizations, then it actually benefits the government as a whole for 18F to do it for free.
While I like that they focus on revenue to prove the impact they are having, selling products to other agencies shouldn't be profit maximizing. Any inefficiencies, either to the buyer or seller, are taxing to the organization.
Side note: $32m is nothing. I'll take that any day over people building GIF hosting and SnapChat.
I smell a hit piece from a mile away. 18F has done a lot of good but in trying to change how government IT does business but they've also made powerful enemies who view them as a threat to their power base..
Find something that can be spun negatively and sell the story to a friendly journalist. That it what appears to have been done here. Next get the head of a budget committee to question 18F spending. It is all done behind the scenes and sadly 18F's strongest supporters are all leaving government.
Now I don't think GDS walk on water, but their work really is much better than how things normally get done in the civil service.
The press focussed on, "This bureaucrat paid market rate for in-demand skills! Disgusting!" Rather than, "In house experts save tax payer being ripped off again!"
That is so depressing. I lived in London from 2011 through 2014, previously having no clue about GDS or anything else in the UK for that matter. Being a ruby developer, it kind of blew me away when I met so many GDS developers and saw what they were doing. They were operating like a top-tier commercial agency much moreso than anything that resembles government in my mind (and I'm half-brazilian, so US government to me is far from the worst in terms of inefficiency and boondoggles). Next thing you know they stop letting the non-white people in, and before you know it Berlin leapfrogs London as the primary tech destination in Europe.
That is a great point in favor of it being a hit piece, because that's wrong - rndmize pointed out below [0] that it's a misinterpretation of the report.
As far as logo design goes, this is actually pretty fast. Most companies I've dealt with stuff like this usually takes much, much longer depending on how picky the founders are.
Divide 3.3 months across multiple people (say 5), and you get less than 3 weeks.
Also worth mentioning is the $X million dollars spent on redesigning Merrill Lynch's logo. The only redesign that occurred was a different angle view of the lion.
I paid $5 for a logo on Fiverr that took the artist about 3 days (general turnaround time, not active work time). To this day, I feel really bad about taking advantage of an artist like that.
$32m is not nothing, it would pay for a big chunk of a lot of programs. The federal government could actually pay the amount of educational money promised in the original legislation to the tribal colleges for many years. It would pay our tribal college's general fund for well over a decade.
Saying "it's an acceptable cost because the $32 million loss is balanced by gains in other areas" is potentially winning argument.
Saying "it's nothing" is a losing argument in every town hall meeting that ever takes place. $32 million is a lot of parks, operations, libraries, and whatever else you like.
We should focus on the value bought by that money, not on waving it away as "nothing".
> Saying "it's nothing" is a losing argument in every town hall meeting that ever takes place. $32 million is a lot of parks, operations, libraries, and whatever else you like.
You have changed the scale by 2 orders of magnitude to make it something.
$100 is a 5th of someone people's yearly salary, so how dare you waste it on entertainment every month?
I left out that I meant 2 orders of magnitude in government. Comparing federal spending to what would be waste in a town or county is an example of giving into propaganda steered stupidity as an acceptable norm.
I don't disagree that this is a terrific value. I disagree with it coming up for discussion like this in absolute terms when nothing the federal government does that is blatantly absurd could ever discussed in absolute terms without seemingly like an unworldly statistic at any local level besides, maybe, in NYC.
There are 34,000 $32 million dollar units in federal discretionary spending. I'm willing to bet that if someone randomly samples another one it will be worse than this one.
That 32 million may be enough for a town to redo it its main library if it received it through complex graft using a rider (and then some kind of magic to get it to the town level) is a bizarre distraction from anything meaningful at the federal level.
I believe the Government should spend the money, but I also believe that you can't wave away any concerns about it by claiming it's not worth worrying about.
Well, I like the "see also" in that entry.. Do you think they will buy a school or a prison with their graft? Will the 99 other senators who want $32m each be so good as your own?
If you want to reduce federal spending and federal tax so that the local system can tax then that's great. If you question the federal government spending $1 out of $34000 on a reasonable global service, then please tell me why my wife would be reasonable in questioning my $1 a year new year's burger, but my converting from a standard $40 a month ATDT service to a $3400 a year program to turn our kids into psychopathic murderers who are sometimes willing to protect us is an unquestionable rational expense.
If you don't pay even $100 in taxes, the government will seize your assets with a levy or - if you continue to resist - send goons to threaten you at gunpoint to turn the money over. Snapchat gets a free pass, because they don't force their investors to pay using a police force. The IRS maintains its own SWAT team dedicated to enforcement of the tax code.
I'm not saying the FBI shouldn't get a TV in the breakroom, but there's no level of money that shouldn't have a good justification.
If you owe $100 in taxes, the government will ask you nicely to pay it.
Then, when you refuse to pay it, they will mark it up with a bit of interest, and ask you nicely to pay it.
Then, when you refuse to pay it, they'll deduct it out of your next year's tax return, and call it even.
Then, if you didn't have a tax return, they'll put a mark on your social security account, and will pay themselves out of your social security cheque.
Then, if you didn't have a tax return, or social security, they'll put a lien on your house, so if you refuse to pay it, they will get paid when you sell your house.
Then, if you still haven't paid your taxes, and don't have a tax return, or social security, or a house, they will complain to a credit agency, and ding your credit rating.
Then, if you still haven't paid your taxes, don't have a tax return, or social security, or a house, or care about your credit, and you are not going to suffer economic hardship because of it, they may take money out of your bank account, or your car, or your property, sell it, and give the remainder to you.
Then, if all those means are exhausted, and you have been particularly belligerent about this process, and owe a lot of taxes, you might be looking at jail time. (At this point, have you considered just paying your debts? Or, perhaps, declaring bankruptcy?)
They will give you many, many, many chance to pay what you owe. The steps that they will take to collect are both reasonable, and proportionate.
Please don't spread nonsense about armed goons at gunpoint stealing 'your' money. That would be civil forfeiture, courtesy of your local police department, not the IRS.
Actually, what you wrote there has some false parts. They send a certified letter and will put a lien on you bank account for the amount if you don't respond fast enough[1]. They add interest immediately, there is no grace period. You can arrange monthly payments (they are required to let you do this the first time). They will deduct the money you owe from the next year's return.
1) Fast enough can get a little weird. Had to loan a friend money because of a mistake and he didn't even get the letter before the lien happened. Saw him pull up his bank account on a Monday morning after his card was rejected at breakfast. He came in the next day with the certified letter and listened to him call the IRS about the problem with his return. They did get new computers a couple years ago, and it was not a good thing since it seemed they might have sent the first letter to a prior address. They were sorry, but too bad.
The context is someone arguing that you shouldn't worry about the federal government wasting $32 million, because it's chump change for them. My point is the polite requests they make are ultimately backed up by these SWAT teams, and that you shouldn't take taxation lightly as a result. Obviously very few cases require them to deploy a SWAT team, since most people would rationally choose to pay the $100.
I'm not saying taxation is bad or that they'll jump on you at the first sign of disobedience, just that the commenter's callous attitude towards spending tax money is bad. It's a very entitled mindset that no government official should have.
>In a perfect market, they would be able to sell their services at the same price as the value of those services.
Uh nope. A perfect market means the price is equal to their costs, not the value provided. A monopoly is when you can extract the value provided as rent.
I don't think they are a monopoly. Other agencies can pay outside firms as an alternative to using 18F, can't they?
I'm not exactly sure how it works so please educate me.
I also separately heard that 18F wasn't supposed to bid against outside firms for government contracts. That doesn't really make sense, then, if they're supposed to earn revenue. If they're expected to earn revenue then they should be able to bid for government contracts.
At least Australia's version of 18F (called DTA) has an enforced monopoly
The Australian Government has established a moratorium that stops Commonwealth
(federal) agencies from investing in new, or duplicating, service delivery
capability. The moratorium is an important tool to help government limit the
fragmenting of users’ experience across the Australian Government.
Yep. Revenue really isn't a useful metric for an organization like this. They may save the entire government money even if they are recording losses on their own balance sheet. The opportunity cost of using private contractors vs. doing the work in-house through 18F is what should be emphasized.
The USDS is working on vets.gov, not 18F. Apparently this effort involves taking hundreds of microsites that all look and behave differently and combining them into one unified effort. It's not surprising that this would take a long time, and my understanding is that this is one of USDS's top priorities.
I think it's easy to underestimate how hard it is to ship something on the scale of vets.gov from the outside.
Source: had early talks with the VA to join this project. I didn't, but came away with a lot of respect for the team working on it.
Right. I am a veteran that uses government portals for certain services, and you are absolutely right. There are so many moving parts behind the "single" portal the government wants veterans to use for access to services and information. I have been using those services for many years and haven't even seen them all, let alone used them all.
I'll give vets.gov credit for mobilizing content with responsive design.
But as expected, "uber-portal" strategies that re-purpose content from other sites quickly suffer from entropy and information rot. The portal just becomes a cut-n-paste curation exercise.
Ultimately, the issue is whether 18F is more effective than other means and accurate accounting is important to that end.
Let's say that the VA gets bids to create a new system for something. 18F says they can do it for $900k and ContractorX says they can do it for $1M. 18F gets the contract and produces a system that cost the VA $900k. Let's say the system cost taxpayers an additional $1M that isn't billed to the VA. That means that taxpayers are out $1.9M rather than $1M going with ContractorX. The issue can be that 18F isn't accurately pricing their services and hiding the actual cost to the taxpayer.
That isn't to say that I like contractors. I don't. But the point of 18F should be to prove that the government can create its own digital services at a lower cost than contractors. If they're "cheaper" in that they're charging agencies less, but then just taking money from a slush fund to cover the fact that they're not efficient, that's not a good thing.
Also, price being "value created - effort to get paid" isn't a good metric. You need to look at alternatives. If a system saves $1M, but you could pay one source $100k for it and another source $900k, you want to go with the $100k source. If we had to pay the value created for things, we'd stop a lot of economic activity. In fact, why would the VA want to do anything at all given that they're going to be charged the full value of any savings? 18F shouldn't be credited with $1M in savings if they spent $900k to achieve it when a contractor could have done it for $100k.
The value of 18F is if they can produce digital services better than other methods. Other methods are generally paying some non-governmental company for the services or hiring direct employees for the agency. In order to determine if 18F is a good model, they need to charge agencies enough to cover their costs. Otherwise, it's ridiculous. Someone in Health and Human Services would say, "we could hire people to do this project directly, but then we'd be paying 100% of their cost, but with 18F they'll charge us a fraction of the employees cost. In fact, we'll tell them to hire the dozen people we were going to hire and we'll only pay X% of the salary rather than 100%".
Plus, if agencies don't have to pay the full cost, they'll do projects that don't make sense. Let's say that a project for the VA creates $1M in savings. 18F says they'll do it for $900k. The VA green-lights it to save $100k. Great, right? $100k for spending on other things. But 18F actually spends $2M on the project. The VA gets another $100k in their budget, but the government is actually down $1.1M. When people see situations like this, they want to take advantage of it since they're able to basically spend other people's budgets. "If you use 18F, a bunch of the money for your project doesn't come from your budget," is the kind of thing higher-ups dream of.
I think 18F will get better at this. Figuring out how much one needs to charge is hard since it's hard to know exactly how much effort a project will take. I'm guessing that 18F is looking to generate good-will with agencies rather than going back to them with cost over-runs. Still, 18F needs to prove that it's a more effective way than alternatives and it can't do that by just under-charging and subsidizing whatever projects they get given. If 18F just offered free services, then every agency would want infinite 18F work (since it isn't coming out of their budget). Then which projects should be worked on? The ones 18F finds interesting? 18F wasn't created to be the arbiter of what the government should do. As such, they need to price accurately based on their costs. They will get better at this. They will likely have to start passing along cost over-runs to agencies. But if they get great, motivated people, they could provide digital services at a lower cost - it just has to actually be a lower cost.
Even if you hate contractors, a lot of agencies have tens of thousands of employees. Health and Human Services has ~80,000 employees. The VA has ~313,000 employees. Why shouldn't they just hire an engineering team? Why should they outsource a project to 18F? Agencies that large certainly have enough engineering projects to keep an internal team busy. They're not some restaurant that outsources their website because they don't have enough work to keep a dev busy. Well, maybe 18F can gain efficiencies through hiring better people at higher pay who can create better code and re-use stuff across agencies. Or maybe the overhead of outsourcing outweighs those efficiencies. That's something important to figure out and it's an answer that we can get closer to if 18F accurately charges agencies.
The biggest issue with this is when contractor x bids $1m to get in the door and then says oh yeah, we need an extra $5m to do it properly since we had more meetings and scope has changed.
None of that is able to be specced up front because people are involved that have "connections" and "back rubbing"...
It's not to say that the extra scope and cost of $5m is worthless, just simply bidding for work and coming in the cheapest is not an accurate view of the work once the project is finished. This is where 18F should be creating value because they are on the Gov't side as were.
> The VA has ~313,000 employees. Why shouldn't they just hire an engineering team? Why should they outsource a project to 18F?
They're working on that! Marina Martin, the CTO of the VA, has been building out the VA Digital Services team with some awesome people. Vets.gov is growing fast, and is open source, and should hopefully help a lot of veterans.
I work with a lot of those people, and they're doing really good work.
"Let's say that the VA gets bids to create a new system for something. 18F says they can do it for $900k and ContractorX says they can do it for $1M. 18F gets the contract and produces a system that cost the VA $900k. Let's say the system cost taxpayers an additional $1M that isn't billed to the VA. That means that taxpayers are out $1.9M rather than $1M going with ContractorX."
If the government is paying itself $900k from one department to another, how are they out $1.9m?
Those $900k are not staying within "the government", they are being spent in salaries, machines, services or whatever. Total government spending would be $1.9mn in that case, $1mn with the contractor option.
18F is another money grab. A bunch of SV people said to themselves, here is an easy few years to make money and if nothing comes of it we can say "Ah well start up culture...lol". I never had high hopes of this, it's sad that it was touted as a startup. 99% of start ups fail over the course of 5 years....
If it's a money grab, it's a badly designed one. I interviewed with 18F and would have had to take a pay cut even on a cash basis. (Government salaries are strictly limited by law.) They certainly don't give stock, and (gasp) you have to buy your own meals. So financially it's not a good deal at all.
They're looking for people with a different motivation than maximizing personal profit. The people I talked to there were motivated by public service, solving big problems, and getting shit done.
Activities that can’t be billed to other agencies are another drain. One example is staff time totaling 727 hours on a logo change. The old logo was a blue square with 18F in the lower-right-hand corner. The new logo, a black square with 18F centered and a different font, wasn’t worth the time and doesn’t look as good.
This reminds me a time when I used to work closely with a design team in a web agency, it was very frustrating to work with them because every little modification to things like colors, sizes, buttons and the like would be accompanied with several hours of "design strategy", time that the designers would waste thinking and prototyping new ideas for the corporate image of any of our clients, I say "waste" because most of the time they would end up with a design similar if not equal to the original.
Their GitHub organization [1] shows significant activity, it is sad to see people spending their time, energy and resources on things that are ultimately not supported by their financial funds.
It took a couple of experiences dealing with serious, fancy pants designers to convince me that most of them are charlatans.
Even something as simple as widening a button was stonewalled by "explorations", "design strategy meetings", "pattern library meetings", etc. The end result was getting even a single component mocked up took at least a week and was no better than what I could have come up with in 30 minutes.
Quite a few UX/design people who have actually gone through education/training (as opposed to learning by doing) seem to suffer by an extreme form of process overload.
It makes sense I guess. It's really easy to teach processes and it probably makes the them feel "professional".
I don't want to blame UX people entirely for that. Designers have often had to elbow their way in to traditional software processes. One way to do that is by looking impressive to executives. It's sort of an extreme version of Calvin's clear plastic binder. [1]
I think it's made worse by the ubiquity of the agency business model for designers. Consulting agencies are forced to learn learn looking impressive and process-y as a vital sales tactic. It's not just design, either. Software consulting agencies have the same problem; if your clients can't directly evaluate your work, then you end up getting judged on what the people hiring you can understand. And that's often how well you dress, how glibly you talk, and how impressive your process seems.
I'm an UX/design person and I notice that too. I think you could put two capable teams to create a website for some company, one required to do a lot of research and documentation, and another left to design it as they please, and the result probably wouldn't differ much.
It's akin to software engineers spending time to automate tasks that aren't that recurring.
This reminds me a time when I used to work closely with a development team in a startup, it was very frustrating to work with them because every little modification to things like functionality, Git workflow, Javascript framework and the like would be accompanied with several hours of "dev strategy" and scrum meetings, time that the developers would waste thinking and prototyping new ideas for the tech stack for their managers, I say "waste" because most of the time they would end up with a system similar if not identical to the original.
From the actual report - "18F Branding – Staff spent 727 hours (valued at an estimated $140,104) developing the 18F brand. An example of one of their branding projects is the 18F logo change, seen below in Figure 8."
Pretty poor journalism. Brand development strikes me as a much bigger category than a single logo change.
Indeed, I remember a few posts/comments by 18f people here on hn - surely those hours taken were "spent developing the 18f brand". Of course, those people might have been doing that work for free, in their spare time.
Meh. I'm sure google spent millions on their logo change, and it was just a font tweak. Hackernews was gushing. God knows what they spend per week on google doodles.
There's no winning here for 18f. Either their sites are archaic disasters and they are criticized for not being up to par with the top tier consumer sites. Or they spend some money to bring them up to par and its "OMG! An $800 toilet seat!"
I want government sites and buildings to look like archaic disasters, because it at least implies that my tax dollars aren't being pissed away on frivolous opulence.
Google can spend ten billion dollars changing their logo because they're profitable and they answer to their board & shareholders. I don't remember anyone asking me if I wanted to contribute money to a government tech startup.
Government sites that look like archaic disasters are like that because the money was pissed away by people who have no desire to learn and want job security. I watched it happen a lot. Tech leads, with tech lead salaries who don't even know simple things like dependency injection and MVC design patterns. Yeah, that's where your tax dollars are going. I'm hoping that I can start my own business and get into this space. There's a lot of money sloshing around, at least I can bring real value and deliver working products while I'm collecting some of it.
"I want government sites and buildings to look like archaic disasters"
And what do you propose doing about people who use arguments of the form: "Can you believe this government $thing? It looks like an archaic disaster. Government can't do anything right. Therefore..."
I don't have any credibility or knowledge to talk about the exact current issue at hand but don't we at hn pine over the beautiful past where email was distributed instead of going between Microsoft, Yahoo, AOL, and Google? A private email server isn't bad by definition, is it? Perhaps we should try to help the "ordinary" person keep their servers secure?
> "Perhaps we should try to help the "ordinary" person keep their servers secure?"
Security wasn't the goal, evasion of oversight was.
Even if it was granted that security was the goal, you would expect the private server to be at least as secure as the @state.gov mail servers, otherwise you're wasting your time. Furthermore, James Comey's statement in July made clear that the private server wasn't as secure as GMail, the kind of email server an "ordinary person" might use. GMail may be centralized, but it's supported SSL from its public release in 2004. The server Hillary Clinton used didn't support SSL until the spring of 2009, months into her tenure as Secretary of State.[0]
The truth is we actually know very little about the operational security of the server because all but one of the IT folks in charge of running it plead the 5th before congress and only spoke privately with the FBI under conditions of immunity. Justin Cooper was the only one to actually testify regarding the specifics of the server, but his role was more administrative than technical.[1]
There's no credible defense for the SoS running a private server to provide email accounts which are less secure than the standard accounts used by the rest of the State Department. The diplomatic role of the SoS inherently requires a high level of OPSEC to carry out its most basic function of communicating with international leaders. Hillary Clinton avoided using the security available to her in order to thwart federal regulations regarding the preservation of public records. That action is no example that any member of the public should be holding in high regard as something worthy of emulation.
What business does a government agency have spending nearly 800 hours (plus inevitable expenses on stationery, biz cards, signage, etc) on a logo? How does this benefit the taxpayer? Why do they need this logo? This logo changes nothing for anybody except the people who worked on it. This is not good stewardship of our tax dollars and you can bet your lunch that this won't be the only FW&A problem found at 18F.
> What business does a government agency have spending nearly 800 hours (plus inevitable expenses on stationery, biz cards, signage, etc) on a logo?
Presumably because they are trying to be a "startup", they are copying the business style of startups. It's easy to look back afterwards and say "that was a complete waste", but it's possible that at the time it was seen as justified. It's easy to say we'll just take the good aspects of startups and the good aspects of government agencies and reject the bad", but it's not always apparent what is the good and bad in each, as they are both fundamentally systems whose ultimate behavior is emergent from all their aspects. It's very easy to accidentally break something useful by tweaking or removing something that appears completely unrelated in systems like that.
That said, I think the more interesting question is should we be trying to emulate a startup in government, given that the vast majority of startups fail? In that light, it should be absolutely no surprise that 18F is failing. It was the likely outcome.
What business does a government agency have spending nearly 800 hours (plus inevitable expenses on stationery, biz cards, signage, etc) on a logo? How does this benefit the taxpayer? Why do they need this logo?
One thing that bothers me about government websites is the boosterism of whoever is in office. Secretary of State websites are particularly bad in this regard. Usually you go to the SoS website to do something related to licensing, like the DMV. Both Illinois and California have had headshots of the Secretary of State displayed prominently in the header. It is labeled/named something long winded like "Secretary of State Joe Blow's Department of Motor Vehicles Website". Is it on a .com or a .gov or a .(state).us domain for any given state? The generic "(welcome and platform) message from the Secretary of State" is often front and center.
Really, who the Secretary of State is is so secondary to the actual services being provided, all this does is serve to tell everyone who is responsible for this crap website.
If there was consistent branding, and that branding was strong and recognizable, citizens would be better served. You wanna toot your horn, put some metrics up like the last 6 months of turn-around time on license approval, website and phone service response time, help topics researched, costs to provide the services, etc, and show me how access to state services are improving over time, which is the SoS's actual job. Government agencies usually have weak branding because it gets upset based on term lengths (except for services like the IRS, which has pretty consistent branding and methodologies).
This isn't just a website problem either - the DMV in Illinois is really the Jesse White center based on the amount of branding and pictures you run into there.
The California DMV website has changed recently, and the headshot are in the footer, as links to their respective sites. This is an improvement, but the branding is still all over the place.
There's probably a legitimate story behind this right? This sounds like the randomizer IBM made for the TSA where the cost was shocking but was explained away by overhead or typical government costs.
From an aesthetic perspective the old one looked like a sign from Studio 54. It was a late-70s throwback font: cool looking, but not really the right connotation for a tech team that is supposed to be dragging the government into the present. The new font and alignment is more neutral and graphical.
From a practical perspective, the new logo retains its legibility better at small sizes on screens, compared to the old logo's smaller thinner font. And making the new logo black and white reduces variants and therefore reduces cost. It's the same logo file whether you're printing 1-color, 2-color, or full-color.
The standard rule is to double the rate for staff to account for overhead - healthcare, retirement, paid leave, etc. If they're paying a GS-15 staffer $125k I'd be surprised if the effective net cost is significantly less than $250/hour – and that's probably a big win compared to a contractor charging you $250 for someone making $90k.
I heard from a few former employees that back in the '00s, a few years before their bankruptcy, Kodak spent close to a million dollars on redesigning their logo. They changed a font.
727 employee hours, in comparison, seems like a pretty sweet deal.
Not to mention that the logo change was only part of that 727 hours (But out of context, it sure makes for an awesome soundbite!)
That's almost unbelievable. This is completely absurd, on par with the mythical $800 government toilet seat, except that this cost us taxpayers far more. I'm completely disgusted. This is textbook fraud, waste, and abuse.
By the way, like a friend suddenly shaving off all of their hair, I consider a corporate logo change to be a harbinger of bad things to come.
> ...on par with the mythical $800 government toilet seat, except that this cost us taxpayers far more.
Sorry to burst your disposable consumer-centric bubble, but not all toilet seat applications are optimized for profitability at economies of scale.[1][2]
Did I misunderstand or is the story also - A government department was started with the ambitious goal of "zero budget" and this year with a 183 person team will have recovered 68.75% of its costs.
And did we forget that fast growing startups don't usually break even either?
I'm currently working for a UN-agency as a contractor, and my team is modelled in a similar way, like a private development agency so to speak (calling us and 18F a startup is a bit click-baity IMO). We only work internally for our agency, not the whole UN, but there are structures like that too - e.g. ICC provide ISP and hosting services to the whole UN. I'd imagine it's a fairly common structure in large companies/organisations.
Anyway, whenever a department wants a new IT project they have the choice of releasing a tender to the public or going to us. We then work like an agency, providing an estimate, then billing them for X days of development.
The first bit where it gets tricky for us, and I'd imagine 18F, is the specs usually aren't clear. Customers decide they want something different, didn't explain what they wanted properly, or something just takes longer than you estimated. Unlike an agency we have to deal with the code and people for a long time, so we can't just say "no you only payed for X days work" and leave something broken.
The other part is maintenance - that is paid for by our internal budget, but as with any software projects there are always tweaks/changes that need to be done. Unfortunately the level of beuracracy is quite high, so getting budget for something that is say 3 days work is more effort than it's worth, so it often just comes out of our budget.
Anyway, even if we lose money I think it's a lot better for us to do this than an external agency. For starters an external agency would be a magnitude more expensive. As we know all our systems we can recommend and build systems that are more integrated with each other, where as an external agency would just build one system in isolation as per the spec. I'd imagine it's similar for 18F.
Our design decisions are Bootstrap for everything, so no we aren't spending a crazy number of hours on just a logo :-)
This is disappointing to read, given I love what 18F has done. No matter whether we think they should be breaking even or not, they shouldn't be creating wildly inaccurate revenue projections and then falling drastically short of them. That looks like something is going drastically wrong with management there, especially considering greater macroeconomic conditions have not changed drastically enough in a way that would cause something like this. I'd hope there is something other explanation though.
The difference is, I can choose to buy a startup's product, or I can not and watch them go bust with someone else's money. This is the government, they take your money by force and you are forced to "purchase" the "product". Because you have no choice, it would be amoral to treat the government like a startup and compel citizens to fund this "product".
It is clear that they have not billed enough to break even internally, but have they saved/made more for other agencies to break even for the government overall?
Presumably (maybe a bad presumption), agencies that use their services save more/make more than they are billed for.
It would be a shame to kill a group that was saving money overall because they did not bill enough internally.
The sniping about 18F seems pretty transparent: federal IT shops don't want to do anything differently, and private IT vendors don't want any more competition for their contracts.
18F might be one of those ideas where you can tell that it's working because it's getting attacked.
One piece of 18F that I've always appreciated is their published API standards - https://github.com/18F/api-standards. They spawned it off the frequently mentioned White House API standards, and it phrases a lot of the opinions well without being preachy.
But more than anything else, this made me feel like there are real humans behind 18F. So I wish them all the best and hope this article (and ones like it) falls into the camp of "Haters gotta hate" and just inspires them to find more visible success.
A surprising number of news pieces actually end up giving you some sort of opinion.
As a made-up example, a news article on Samsung's falling stock might focus on its Note 7 failures, making the assumption (and thus giving the reader the impression) that the mobile phone is the main cause of the company's decreasing revenues, when in fact the company's general electronic components sales have been dying for months. But this slow decrease is much less sensational and might've missed the headlines. This sort of "opinion" is far more difficult to notice, but still exists in reporting in the form of attempting to speculate on the reasons behind the facts.
This article just happens to be particular blunt about its assumptions.
> This sort of "opinion" is far more difficult to notice, but still exists in reporting in the form of attempting to speculate on the reasons behind the facts.
But that’s how it’s usually marked – as a seperate, independent part of the article, marked opinion or speculation, or by putting it in context.
Hi Silicon Valley! Thanks for visiting our nation's capitol!
kisses,
DC
--
Seriously, this shouldn't be surprising. It's a noble effort, the work is probably worth the so-called "losses", but the reason the government is the government and not Silicon Valley is that in government, there are a bajillion rules to keep you from screwing up too badly.
There's been a fetishization in DC of Silicon Valley techniques, but 1) it's not at all clear that the current Silicon Valley ethos really is worth replicating, and 2) the conditions aren't right here, it can't be replicated without fundamentally changing the conditions. One of the conditions is that there's oversight over everything, and there's the press.
One little article in the federal section of the Post isn't going to kill off anything alone, but it will be interesting to see what happens to 18F after the presidential transition. If there isn't high level cover for it, it will be ground to dust.
> it will be interesting to see what happens to 18F after the presidential transition.
If Clinton gets in, she'll probably appoint people from Google into high government positions. Eric Schmidt has given her campaign a lot of money and technical support, and she's mentioned bringing people from Silicon Valley in during some of her rallies. Probably that means giving Eric Schmidt a high-ranking position, or Google lucrative government contracts.
I don't think that Trump cares too much about Silicon Valley. He'll probably just review them as part of a routine audit of all federal agencies, and ask them to explain what it is that they do and then make a decision to keep them. I think this particular agency is not high on his list of problems.
Even though many people see "guys" as a harmless, genderless word, not everybody does. You can think of the opposite word to guys (for example: gals or girls) and wonder if a group of men would feel comfortable being referred to as girls. When someone refers to you using a word that you don’t identify with, it's easy to feel excluded from conversation or misidentified.
Ha ha ha this Americans and their political correctness. I cringe every time I see things like this and the HN community doesn't escapes from it, in fact, I have seen more "political correctness" towards sexism here more than in other communities on the Internet (at least the ones I visit). I still do not understand why people pay so much attention to these things. I can only imagine how people would react in countries with a language with Latin roots like French, Portuguese, Spanish where there are multiple ways to refer to a individual or group of people with different genders more than in neutral languages like English.
> I still do not understand why people pay so much attention to these things.
Quite a lot of people who respond as you do actively don't want to understand. But in case you're one of the ones who is actually curious:
Historically, sexism has been a problem. For millennia, women were basically property. The last couple hundred years, we've been unwinding that. E.g. less than 100 years ago, women finally got to vote. They didn't have equal access to credit cards until 1974. In the 1990s, US states finally all made marital rape illegal. In the early aughts, women reached parity in the proportion of people getting medical and legal degrees. (The proportion of actual doctors and lawyers hasn't caught up yet; it's around 33% now, and I'd guess it will hit 50% about 2040.)
One place that sexism still resides is in language. That language at least sometimes shapes thinking. Many women in professional contexts reasonably object to gendered language that excludes them. So to be polite, many workplaces are shifting to use more inclusive language.
So, TL;DR: we're in the middle of overturning millennia of pervasive sexism, a language shift is part of that, and if we're lucky we'll be done with the project in only decades.
Regardless of what speakers of those languages do, we can do everything we can to remove gender bias from English. We don't have to keep doing something just because others are doing it.
Yup. I don't have a solution for gendered languages because it's not a problem I deal with much. I'm going to work on English, and I'll trust that other people will find the right solution for their cultures.
This isn't about political correctness, it's about using the words you actually mean. If you are addressing a group of people, it makes sense to include them all in your address. So use a word that includes everyone, as the bot mentions, there are plenty.
If you think "guys" is gender-neutral, how about a thought experiment. You're a man, you started dating a woman. Do you tell your friends, "hey, I just started dating this great guy!" Nope. So why use that as your generic pronoun for your coworkers? It unnecessarily makes people feel excluded, even if you don't intend to exclude them.
100% on board with this. But I think the "dating a great guy" example is slightly off. In many languages, the masculine plural refers to a group of all men, or a mixed group. Whereas the feminine plural only refers to a group of women. And of course masculine singular refers to male only, and vice versa.
Saying "guys" in English refers to a masculine group or a mixed-gender group historically. Same with Spanish, for example. The suffix -o is masculine and -a is feminine. But -os is plural masculine or mixed gender group, -as is feminine only group.
That's generally just how language works. When I personally speak to any group (or person) I don't use gendered language, but although I like your example, I don't think it holds up to scrutiny.
Right, "man" being the default gender is somewhat unfair to half the population, so it's best to avoid it. Yes, this means you have to think of new words. Nobody had a problem when their phone keyboards were too shitty to type "you", so I think we can do it.
So, go back to the beginning of the comment thread; it's about how a bot that only speaks English is "dystopian". English doesn't have gendered words, so we can quite easily fix the problem with almost no effort at all.
"Guys" is plural for "guy", which literally means "man". The plural has informally been made gender neutral in an attempt to be inclusive, but the connotations will always remain given what the singular means, so it's a pretty regressive attempt at inclusivity.
Last time I checked "guys" was a plural for "guy", which literally comes from the British, meaning, "a person of grotesque appearance" in relation to an effigy Guy Fawkes or those who dress to resemble him on November the 5th (essentially British Halloween). A word to which Americans only gave masculine overtones much later. So while the singular has been informally made gender specific, the plural is--and has always been--gender neutral as both sexes dressed as "guys".
That's some very creative selective historicising.
The modern definition of guy is man. Unequivocally. Beyond the very seasonal references to the effigy in England, that definition is not in general discourse outside of tomorrow's holiday and your archaic definition is certainly not used by anyone save etymologists.
The modern defining to mean man is not purely an Americanism by any stretch.
The pluralisation is also contemporary.
Regardless of etymological technical "truths", can you really say with intellectual honesty that there's no masculine connotation inherent in "guys"?
"Regardless of etymological technical "truths", can you really say with intellectual honesty that there's no masculine connotation inherent in "guys"?"
I'm sure there is to certain people. But to me, my family, friends, and coworkers I have not found it to be the case, no.
In fact, I hear female coworkers use the term neutrally every day. We as a team even use the term over conference calls with people all across the country and I have yet to hear anyone take issue, offer correction, or purposefully use alternative terminology.
I simply don't see this as a problem, nor can I understand why it needs to be purposefully corrected. It's the same PC attitude that causes people to take issue with those who call African Americans black. To me, black is a more useful term. If it wasn't in my repertoire, I'd have no way to distinguish Idris Elba from the rest of an essentially all white cast ...
Here's a fun test. Tell me which gender connotation the following terms hold: Cheerleader, Usher, Flight Attendant, and Comedian. If you said female, male, female, male, you'd be wrong as all are gender neutral.
They're essentially an internal dev/consulting shop and they needed to approach it from this stand point or they'll continue to be underwater. Every contract that IBM or Accenture, et al wins, 18F should have made the sale that they'll be there guarding the hen house as the project progresses - easy sale and easy project staffing. Branch off from there and take down small dev jobs to build mass and it'll snowball from there if managed properly.
Leadership heads need to roll, ranks thinned, and they need to focus on the mission.
So they should nickel and dime them on scope changes and deliver solutions that meet the requirements but are useless? I know I'm straw manning a bit here but asking them to operate like these contractors is highly questionable. Their mission isn't to be cheap and competitive but to change how the government uses technology and make it more agile. To drive culture changes and set standards that allow government agencies to produce solutions that are aligned with their goals.
How many failed government IT projects have we read about that fulfilled a contract but was ultimately useless? Or stories about government inefficiencies and a failure to keep pace with modern technology?
I don't know why there's an expectation that they do this cheaper or at the same price as government contractors. Have we forgotten what happened with the healthcare.gov site already?
I love 18F! A few months back they published a set of web design standard [0] that I really liked. Accessibility is among their top concerns, so it's legible in all devices and there's a lot of contrast.
I used their color suggestions for my blog [1] (still haven't updating syntax highlighting, terribly sorry about that). The two things I changed were typography sizes and fonts. I decided it would be best to use preinstalled fonts instead of loading a third-party font. I don't think third-party fonts justify slower load times for a static website like my blog. But overall, I really like their color scheme, as it's very comfortable to read on any screen.
Doesn't it seem wasteful that 18f spent all that time on the standards, instead of maybe a guide/theme on how Gov't sites should adapt, already open-sourced (bootstrap), frameworks?
Take a look at https://github.com/18F/web-design-standards/blob/staging/LIC... and count the number of existing open-source projects which they're re-using. This looks to me like an experienced team who did their homework to assess how much flexibility and control they needed and reused as much open-source as possible while still meeting their other requirements.
Frameworks and standards like that are more often needed because there's a huge ecosystem of custom, proprietary libraries and frameworks that also exists within government. Previous attempts to try to release standards that only address specific frameworks that are popular within the open source world may only ever apply for 20% or so of federal agencies. There's even "agency-specific" forks of open source software that looks nothing like what they were based upon now. But really, you have tons and tons of different frameworks out there that have wasted taxpayer money delivering nothing material in value except "the government owns its IP" meaning "it avoids political / legal problems." See https://github.com/ozoneplatform/owf-framework for an example
Second of all, nothing precludes them from releasing standards-compliant variants of frameworks after the higher level standards are complete.
> “Since its launch in March 2014, 18F has struggled financially,” says GSA’s Office of Inspector General (IG), citing loses every year, now totally over $31.6 million.
And from a random site that seems to agree with a few others:
> In fiscal year 2015, military spending is projected to account for 54 percent of all federal discretionary spending, a total of $598.5 billion
In 3 years 18F has spent 31.6 million. Even if it lost 31.6 million every year, it could do that for NINETEEN THOUSAND YEARS and still not equal what we spent on the military in one year.
Imagine a country where 18f had a budget of 600 billion dollars a year.
It's about what an organization spent compared to their budget. It sounds like 18F was budgeted $0 since they were expecting to earn back their costs in revenue by charging other agencies for their services. Since they did not fully recoup their costs, they "lost money".
If you budgeted the military $1 trillion and they spent $1.1 trillion, then they would also have lost money.
I will take cynicism out of this equation and still disagree. Highlevel the ability to strike & defend the country (not simply project) is hugely important. However, having a technically proficient group building tooling to increase efficiency, speed, transparency and collaboration-- not to mention technical security, is certainly important. So yes, I would like a more efficient government && hope we can leverage technology to improve it.
Is 18F delivering on this? I don't know, but it is conceptually much more important than 1/19000th of military spending.
They could've created feature/bug bounties and/or grants for open source contributors, and tried to build an open source community around what they're building. This would naturally attract/increase volunteer contributions, but they might also be able to control costs better with such a model.
18F is an uniquely special position since they are building software for the general public, and so are in a great position to attract OSS contributors, as contributing to 18F would be a way of helping your community/country, and would make people feel better.
One example is staff time totaling 727 hours on a logo change. The old logo was a blue square with 18F in the lower-right-hand corner. The new logo, a black square with 18F centered and a different font, wasn’t worth the time and doesn’t look as good.
I think the Washington Post pretty badly mischaracterized the original findings, which were as follows:
"18F Branding – Staff spent 727 hours (valued at an estimated $140,104) developing the 18F brand. An example of one of their branding projects is the 18F logo change, seen below in Figure 8. "
It was 727 for all branding activities and the logo was just one of them. And that's over a period of almost 2 years. Doesn't strike me as unreasonable, especially in the early stages.
727 hours is a team of 10 working for less than 2 work weeks. Or a team of 6 having a couple 1-hour meetings with other groups in addition to doing the design work. It'd be hard for any medium to large organization with an internal design team to rebrand without using hundreds of hours of staff time.
that said, assuming you are working in the US, are you willing to work an extra few minutes over the course of the next year to generate the tax revenue for this?
> What I don't get is 727 hours for that specific logo.
It's not 727 hours for the logo. It was that many hours for the entire rebranding effort, including coming up with the brand and implementing it. When coming up with logos and brands, it's often necessary to brainstorm ideas, try many of them out, review them with stakeholders, before finally settling on one vision and then executing it.
Not sure how it happened at 18F, but in my experience working with "serious" designers, this is not surprising. They will spend weeks interviewing "stakeholders" and customers, iterating on "visual languages" and "brand identity", etc., etc., going through endless tweaks and revisions and feedback sessions before finally settling on something that looks like a totally untrained graphic designer crapped it out in five minutes. (Obviously this doesn't apply to all designers; there are many who are much more pragmatic.) The recent redesign of the University of California logo is a good example of this phenomenon: http://99percentinvisible.org/episode/episode-69-the-brief-a....
Honestly that's a dumb one to cherry pick, because we only see the end result. If you figure a team of 6, for instance, working 30-40 hrs a week on branding for about a month, the time is easy to figure.
What's harder to understand externally is that there may have been 10, 20, or 40 logo designs that get explored, scrapped, edited, changed, saved, reverted... and sure, sometimes you end up pretty close to where you began. But that doesn't mean it was wasted time or effort.
Design often gets shit on when all you see is the beginning and end. The value is in the middle, which is done internally.
This brings back so many memories from working at a design-heavy web consulting shop during the first bubble. We had some of the best designers I've ever worked with, who could meet with the client once and produce multiple great concepts the next day.
None of that mattered with some clients: get a room full of VPs, an indecisive founder, etc. and before you know it the endless tweaks and revisions blue well into six figures and start delaying work on everything downstream.
Some of that was indecision, poor management, etc. but much of it was simply many stakeholders with different priorities. Get a dozen Sony executives in a room – which was the average meeting – and unsurprisingly it turns out that you have a dozen opinions about what the most important elements of the project are. Someone visiting that website or using a product has absolutely no way to know any of that and sees a design they could do quickly – if they'd only been presented with the end vision of what the project was about…
> The value is in the middle, which is done internally.
Wait. Can you back up and explain that? From the way I think about it, design is about producing something in the end. If you can produce the same end product cheaper and or faster, then you are ahead. If I buy two manufactured items bikes that are physically identical, but one company found a way to reduce the work required by 10% to save themselves, money, that doesn't matter to be as the end recipient of that product. If it's identical, it's identical.
Now, if you're a design firm, sure, the knowledge gained from the work might yield benefits down the line. But 18F isn't a design firm. A new logo does not help them make better products for the agencies they serve. In what way are they better off by spending that time than if they were able to literally just slap the same end result together in five minutes?
I could see 200 employees with a a single all-hands-on-deck meeting where nobody can keep it together and it's three hours of chaos where nothing gets accomplished, but everyone racks up a free lunch, and then a few smaller meetings over multiple weeks that account for the rest of the time.
> Activities that can’t be billed to other agencies are another drain. One example is staff time totaling 727 hours on a logo change. The old logo was a blue square with 18F in the lower-right-hand corner. The new logo, a black square with 18F centered and a different font, wasn’t worth the time and doesn’t look as good.
I thought newspapers were supposed to report facts, and leave opinions to either the op-ed pages, or quotes from sources. Who made Joe Davidson a design critic?
IT is a cost center in most organizations, not a profit center. 18F serves a similar function - it streamlines other organizations, with the hope being that at the end, the cost savings in those other orgs is more than the cost to run 18F. As long as we are saving money overall, then less taxpayer money has to be shelled out to keep the country running, and it is a win. Or am I missing something?
I assume their financial model is to charge agencies for their services. Often times by competing with private vendors. If such an agency is 10 percent over budget, it means they're undercharging agencies by that much. By failing to charge their true cost, they're potentially undercutting vendors who would have cost the taxpayers less overall.
It sounds like upper management doesn't want to do the hard work of accurately pricing their services, or maybe is intentionally undercharging to ensure they get to provide the service. In some ways, they're definitely in a bind -- if they undercharge, they're crowding out private competition, and if they overcharge, they look like greedy bureaucrats.
I have a unique perspective here, as I helped run a .gov (although not at the Federal level) that aspired to run as a self-supporting business.
It's difficult to do because the government isn't a business, and the decision cycle isn't the same as a commercial. Government is appropriation based, uses cash accounting and loves capital expenses. Businesses are diametrically opposite.
There's also a weird calculus for what factors into an ROI calculation and what doesn't. If money is appropriated, there's little incentive to reduce spending (and incentives to spend unnecessarily!) and rate structures don't influence decisions.
In short, I'm not supportive of government waste, but it's pretty clear that any money appropriated to 18F is delivering a positive return. The OMB people should address the perceived shortfall by just taking money away from targeted programs.
"The new logo, a black square with 18F centered and a different font, wasn’t worth the time and doesn’t look as good."
LOL now WaPo writers are also design critics. There's no citation for this statement, it's literally just the author talking shit in an ostensibly "news" article.
The issue here is showing the world how it is possible to be in government and still produce modern open software development.
It is about changing the game not making a profit.
Government software should be all open source (http://www.oss4gov.org/manifesto) but it should also be developed using modern best practises - which is what the GDS in U.K., 18F in the US are trying to prove - they are trailblazers, not profit maximisers.
Looks like some big bloated money-wasting federal contractors that are seeing their margins hit by 18F caught the Post on a lazy day and had their Press Sec feed the Post a pre-written piece so the reporter could get to happy hour early. I expect better from the WP.
Two questions: How much do they pay? Where are they located?
That should be enough to determine if they have any chance at success.
---
I worked for government projects in a small European country. It was actually a decent job in a decent company. I could recommend to people to work there. Salary and perks were average for the area, shit for the world (not much room to do better in the area, no great companies).
IMO: In the major USA locations, this has absolutely no chance to work. There are numerous tech companies who will offer better pay, better work conditions and better perks. Anyone will a brain cell would just go to a tech company and avoid government like the plague.
And if someone wanna try anyway, he'll quickly realize that all his peers are elsewhere, and he's gotta leave to join them.
Whether or not it's worth the money, many true startups that lose that much money die out. The threat of dying off if the company can't figure out how to get sufficient revenue is one major source of creativity in startups. So, it's just another government agency.
The real question is if the start up model will extend to them going bankrupt and being shut down if they don't reach at least break even in X amount of time. If not then they were never really a startup model they were just a standard bureaucracy.
Governments can never provide value like this, government is not a business! Sometimes this trade off is worthwhile. I'd like to have my housefire put out in a timely fashion. I will sacrifice some of my income for this. However, the idea that the government can become some kind of silicon valley startup is just absurd. I can't believe as taxpayers we are forced at gunpoint to fund these ridiculous ideas. And people wonder why we are 20 trillion in debt and our currency is being devalued every year.
This is actually intended as a cost-saving measure. Government IT is enormously snarled. Quite a lot of money is spent on outdated systems and giant contracts with enormous overruns. Very few people with serious technical chops actually want to work for the government as a lifetime career choice.
The theory of 18F is that you bring in high-grade technical talent for 2-4 year stints. You have them jump in to help unsnarl knotty problems, using approaches that are obvious to we Silicon Valley types but not widely adopted in government IT circles.
I know all this because a couple of years back I interviewed with them. I decided to stay in private industry, as I didn't want to leave SF. But they have a lot of really smart and dedicated people who want to change the way the US government does technology, making it faster, cheaper, and more effective.
So I think the real question is not, "How much was spent?" but "How much did they save?" That's a harder question to answer, but I think it's the only real way to evaluate 18F as an initiative.
I believe the idea is less that "the government can become some kind of silicon valley startup" and more that "the government can get tech projects done without involving outside contractors who deliver shit product and charge absurdly high consulting fees"
The big consulting firms have zero interest in encouraging good quality development. They will happily sell you another SAP/whatever system that has a few features that match whatever some set of inexperienced managers say they want, and then spend the next five years customising it for every request.
They get paid on the number of hours, and the number of hours goes up the more complicated something is, so they have zero problem with the number of stakeholders and meetings and revisions going up, because they bill all of that back.
A department like 18F can potentially have a leadership role and say that things should be developed in certain ways (like API standards) and using modern tooling that works across a wider range of devices/browsers/etc.
LOL! Who was expecting different results ? Fatcats in DC use our money to gamble over programs and lose the money. This is not the first nor the last in fact this is the template in which major plans like Obamacare have been molded.
In a perfect market, they would be able to sell their services at the same price as the value of those services. So if they build a system for the Veterans Administration (VA) that saves $1 million, they should get $1 million for it. Their EOY revenue would reflect their value created.
But they are actually selling to themselves. They are a part of the federal government, and their customers are other agencies. So the price they should charge is actually (value created - effort to get paid).
If they waste $500,000 of man-hours to get that $1m, then the government loses $500k for no reason, other than for 18F to be able to point to $1m in revenue as proof of services provided.
The government as a whole also loses out if price is an issue for the buyer, but a net win for the whole organization. If bureaucracy means that a department isn't willing to spend $1m to reap $2m in savings, because maybe the benefits are nebulous or internal politics or any of the other crazy things that people do in big organizations, then it actually benefits the government as a whole for 18F to do it for free.
While I like that they focus on revenue to prove the impact they are having, selling products to other agencies shouldn't be profit maximizing. Any inefficiencies, either to the buyer or seller, are taxing to the organization.
Side note: $32m is nothing. I'll take that any day over people building GIF hosting and SnapChat.