Category Archives: Non-Technical Commentary

Stuff that’s not about technology.

A Quick Review of Mint.com – from both a user and programmer perspective

Posted on by

For those of you who are unfamiliar with Mint.com, it’s a free online personal finance site that allows you to track spending across all of your bank accounts and manage budgets and financial goals. Once upon a time it was a nice little Web 2.0 startup, now it’s owned by Intuit. The service is free, but it’s ad supported, albeit in an unintrusive way – “partner” services (credit cards, savings accounts, etc.) are recommended. I’ve been using it for about two weeks now, with a goal of tracking my spending, developing a budget and getting a better handle on my present and future finances. So here are some of my impressions, considering that the service is still growing but has been around for a while.

The Good:

  • Account Sync – Some things I’ve read online suggest that Mint now uses the Yodlee service to sync transactions from your financial institutions. Whatever they use, they seem to be able to sync transactions and account balances from all of my accounts, even ones at a few small investment companies. The sync is fast, and they do push notifications (low balance, large deposits, etc.) very well. They even seem to support many more institutions than Quicken last time I used it (2010 I think).
  • Budgets – Mint has a very nice, clean, simple monthly budget implementation. You tell it how much you want to spend in a given category in a month, and it shows a nice horizontal bar graph of how far through your allotment you are. It even makes some suggestions, and sets baseline budgets by category based on previous spending. About the only problem is when my bi-weely pay checks fall on odd weeks in the month. Most of my bills are due near the end of the month, and if my pay check happens to be around the same time my bills are due, I’ll have quite a bit less cash on hand at the beginning of the next month than the budgets reflect.
  • Categorization – they do it very well. One of their big selling points is their categorization engine, and it seemed to get about 80% of my purchases right, from day one. That’s much better than anything else I’ve seen. The only down side is that they only look at the Vendor name, not the description – so when I use my debit card and the vendor name is “NYCE Transaction” but the store name/address is in the description, Mint is hopelessly lost. They offer rules to put transactions in category, but only exact matching on the vendor name, so there’s no way to tell it to look at a description that includes “123 Main St”. From a programmer’s perspective, they should allow rules based on either vendor name or description, and maybe they should let the user highlight or specify the part that matches (or a RegEx?).
  • Apps – I haven’t tried the iOS ones, but the Android app is pretty nice. It gives a nice quick picture of your spending and account balances, how you’re doing with your monthly budget, and any alerts/messages. Transaction history isn’t easily accessible and is per-account only, so it’s hard to answer “hmmm, how much did it cost last time I was here?” or “where did I get lunch last Tuesday?”, but it’s still handy enough to tell me how much I have left in my Entertainment budget, or whether I should put this big purchase on my credit or debit card.
  • Goals – Mint has a good idea with goals – it includes a number of pre-configured financial goals like paying off credit card debt, saving for retirement, or stashing away some emergency cash. But in order to function correctly, you must have a specific, dedicated account for each goal. I guess this is where Mint makes money – they suggest accounts with specific attributes (high interest, high liquidity, whatever) from their partners. But there doesn’t appear to be a way to set up a Goal without associating it with a dedicated account, and (as per some forum posts I read, I haven’t tried this) you can’t associate a cash asset account with a goal, so there’s no way to track my progress on building the Depression fund I keep under my mattress.

The Bad/Room for Improvement:

  • Customer Service – I don’t know when Mint was bought up by Intuit, but to me, they have the feel of an agile Web 2.0 property that was bought up by a slow-moving bureaucratic company unable to keep up with customer demands or the amount of support requests. They use GetSatisfaction, a new-age forum-based customer “community” provider, for customer service. But their forums are filled with feature request threads that have “we’ll bring it up to Product Development and get back to you” responses over half a year old. There appears to be very little follow-through or feedback on feature requests, and poor announcement/communication surrounding the one feature that was recently brought in to an invitation-only beta. Unfortunately, I get the general feeling (and I’ve found this far too many times lately) that they were originally a cutting-edge, agile, fast-paced, “cool” service, developed by a team that probably stayed up all night coding a cool feature their users were asking for, that’s now been swallowed up by a behemoth that wants a 6-month-long risk assessment before writing a line of code.
  • Cash Tracking – I understand that Mint makes money from recommending the services of lenders, banks, etc. But if you want to monetize every single feature, you’re bound to fail. In this respect, Mint’s total lack of cash accounting is a serious failure for many of their users (and the forums are filled with threads to this effect). The site allows users to manually input cash transactions, but that’s about it. You can deduct the amount from an automatically-synced ATM withdrawal (just your last ATM withdrawal, no choice if you have multiple ATM accounts, or you and your partner/spouse both use the same Mint account), but that just complicates things further. There’s also no logic of tracking cash on hand – you can create a “Asset” (cash) account, but you have to manually update the balance. If you’re like me and use cash for small purchases, or often get meals at local places that only accept cash, this becomes quite a bit of a hassle. Even more so if, like me, you’ve ever lost an ATM card on a holiday weekend, and now keep some actual cash in the house. Honestly, I think this is my biggest complaint about Mint, and the one that’s most likely to keep me looking for alternatives. More importantly, this has been requested and voted for over and over on their support forums for over a year, with no informational response. For a company whose developers aggregate and analyze data from thousands of banks, I’d think that letting users type in a series of dollar amounts and then totaling it up would be pretty easy.
  • Scheduled Transactions – On the plus side, Mint parses the “next payment due” date out of credit card and loan account information, so it gives you nice (email, SMS, and web) alerts a configurable number of days before. However, it doesn’t even have a feature to set calendar reminders for when bills are due, so I’m still stuck opening Mint and Google Calendar if I need to figure out whether I can make that big purchase today or if I should wait until my next pay check. This was also a very commonly requested feature on their forums, and apparently they have an invitation-only beta running. But I responded to the thread a few days late, so I wasn’t given an invitation. As far as I can tell, there was never an announcement of the beta, and I never got a chance to check a “please let me beta test” box.
  • Forecasting – This is my #2 biggest issue with Mint, after the lack of cash tracking. And the other one most likely to drive me either to another service, to write my own software (again), or to go back to a spreadsheet. Mint is great at tracking your transaction history, your account balances and net worth, and your current monthly budget. But it has absolutely no concept of financial forecasting or balance predictions. If you’re on a tight budget, this can be a problem. At the moment, I have a spreadsheet were I keep a list of all of my regularly recurring expenses (in date order over the month), my pay dates, and the combined balance of my checking/savings accounts. I can easily see what (within reason, not counting discretionary expenses) my cash on hand will be at any time in the month. This seems like a no-brainer necessary feature for a site like Mint, and combined with its great budgeting features, could provide a wonderful picture of what to expect in the next month or two, and when I should plan on making that unusually large purchase.
  • Couples/Families – My girlfriend and I both use Mint, and have somewhat combined finances (but not accounts). It doesn’t make sense for us to have two Mint accounts, since it would be a real pain to keep track of. But, on the other hand, having all of our accounts in one Mint isn’t an ideal solution either. I can categorize transactions, but there’s no easy way to track who the purchase was for, short of duplicating (or triplicating – Me, Her, Both) all of the categories. I’m trying something with tags, but it’s a real kludge in Mint’s search interface to get a list of transactions filtered by category and tag. Also, the account name doesn’t show up in the transactions list unless you pull down the Detail box for a transaction, so it’s troublesome to look through a list and sort out whose cards everything was on.

Miscellaneous:

  • Passwords. In order to sync data from my banks, I have to supply mint.com with all of my login credentials. I understand that, until banks enter the 21st century (and be smart about it – even a less likely possibility) this is just the way it is, and I can’t blame Mint for a system imposed by the banks. But still, I’m not too happy about the idea of Mint storing in plain text (I assume, since they need to pass them on to the bank) my username and password for all of my banking sites. I got some comfort knowing that the site is owned by Intuit – a company that, if not totally security-minded, is at least big enough not to fold overnight after a breach. Maybe one day, banks and credit card companies will let me specify separate credentials specific to third-party software and websites, which I could then revoke with my “master” password. Until then, I guess this is just a risk I have to live with.
  • Importing Transactions – Mint only supports importing up to 3 months of transaction data per account. Some institutions provide less than that. Obviously Intuit doesn’t want people switching from their paid software, so the two YEARS of transactions I meticulously categorized in Quicken are now useless to me in Mint. However, I guess I can’t really fault them too much on this one. Firstly, that would be a LOT of data for them to store. It would also be much more of a headache to try and ensure that the export formats of whatever software packages they support are properly parsed.

Conclusions:

Mint does a lot of things well. And it’s a nice clean interface. But there are also some key features missing, which I feel like I really need. So, for the time being, I’m going to keep using mint, but also explore what other options are available, and look into anyone who offers an API for pulling transaction data that might actually be powerful enough for me to use to write my own web-based app (since my attempts at pulling OFX feeds only worked for about half my accounts). If you don’t care as much about tracking cash spending/cash on hand, and are financially stable enough that you don’t need to forecast the next month or two, Mint is wonderful.

Federal Government: Ignoring what we learned 5,000 years ago

Posted on by

Being a volunteer EMT, and at times an outdoorsy-type person, I occasionally get emails from friends about disaster-related topics. Especially in the post-9/11 days. Recently, I got an email asking me if I’d seen the Department of Homeland Security’s disaster preparedness site, ready.gov. I’d heard a bit about it, but hadn’t really looked around. The site’s main premise is to educate people about how to prepare for a disaster – “get a kit”, “make a plan”, “be informed”.

The site’s own About page states that it is “designed to educate and empower Americans to prepare for and respond to emergencies including natural and man-made disasters. The goal of the campaign is to get the public involved and ultimately to increase the level of basic preparedness across the nation.”

The first step on the site, “Get a Kit”, includes a list of supplies to have on hand as a basic disaster kit. The introduction to the page reads:

You may need to survive on your own after an emergency. This means having your own food, water, and other supplies in sufficient quantity to last for at least three days. Local officials and relief workers will be on the scene after a disaster, but they cannot reach everyone immediately. You could get help in hours, or it might take days. In addition, basic services such as electricity, gas, water, sewage treatment, and telephones may be cut off for days, or even a week or longer.

.

I will admit, I assumed this would be geared more towards urbanites, not those of us in the suburbs or rural America (the latter generally being more accustomed to self-sufficiency), so I adjusted my expectations accordingly. At first glance, it seems to include most of the things that the EMT in me would want it to:

  • Food and water
  • First aid kit
  • Wrench or pliers to turn off gas line
  • Prescription medications
  • Bleach (as a disinfectant and for water purification)
  • Fire extinguisher
  • Matches

However, I found a number of things obviously missing. Sure, some are minor – matches but no mention of fuel, bleach but not even coffee filters to remove particulate matter from water. I also find it comical that so many years after the Cold War “Duck and Cover” approach to nuclear safety, the government is now telling people to Shelter-In-Place in the event of a biological attack – by covering their doors and windows with plastic sheeting!

But by far the most egregious omission from the list – something learned by man at least 5,000 years ago, and taught to every Boy Scout – is the simplest tool: a knife. Yup, that’s right, the Disaster Kit list doesn’t include any variety of cutting implement. I don’t know if whoever wrote up the list is just scared of getting sued when little Bobby finds the knife and kills himself, or whether they’re really that stupid. But I, for one, never leave my house without some variety of the oldest real tool devised by man, and can’t imagine anyone thrown into any sort of “survival” situation without at least a rudimentary way to cut, scrape, pry, etc. I’d say that if I were without utilities or organized public safety for “at least three days”, I’d sure want to be able to make simple repairs, fashion simple tools, or perhaps cut something.

On another (albeit less important) note, there are still many, many Americans, not just in rural areas, who have a fireplace or wood stove. What about a few days’ worth of wood?

Priorities

Posted on by

Why is it that Rutgers cancels classes hours before a football game, and surely would if there were classes during Super Bowl times… but both of my classes were scheduled to meet on election night???

Switch Shopping; the Law of Gross Tonnage

Posted on by

Well, I’m shopping for a new switch. But first, a small aside:

I drive a Ford F-250. It’s a pretty big truck. And every once in a while, someone cuts me off in something like a Honda Fit. It really makes me stop and wonder – how stupid are they? I’ll admit that after two and a half years of driving a 13,000 pound ambulance in my spare time, I know a bit about the maneuverability problems of large vehicles. But I have to assume that the average driver, when confronted with something like my truck – 6500 pounds empty, 21 feet long and tall enough that the hood is at the level of the roof of most passenger cars – they’d think twice about doing something stupid. I won’t even mention the people who cut off or tailgate dump trucks and semi’s.

I’ve spent a bit of time on the water throughout my life. Piloting a boat isn’t especially difficult, but there are a few things you learn about navigation. One of them is affectionately known as The Law of Gross Tonnage. It’s really simple. Big things take a long time to stop, and can’t maneuver very well. So stay out of their way. So I have to wonder, why don’t they include this in driver’s manuals? Sure, people do stupid things. Some people just have to, it’s part of their nature. But if they’re going to, why can’t they be smart, and choose to do it in front (or behind) of a vehicle that actually has a chance of stopping?

In other news, I’m starting to shop for a new switch. At the moment, my network at home (hosting this blog/site and everything else) is running on a single, ancient, Rutgers University surplus Bay Networks BaySTack 450-24T. It’s served me well (and it server Rutgers well before me) but it’s time for an upgrade. While it’s a manageable switch, it’s only capable of 100Base-TX. And streaming a few gigs of backups over it every night isn’t fun, not to mention attempting a network install of a full OS.

So, it’s time to shop for a new switch. Depending on budget, I’ve identified two possible configurations:

  1. Get a new 10/100 manageable switch for the main network, and buy a desktop 8- to 12-port Gigabit Ethernet switch for a dedicated backup LAN.
  2. Bite the bullet and buy a Gigabit manageable switch.

If possible, I’m leaning towards the second option. My budget puts me in the market for what are normally considered edge switches, and Gigabit is just arriving in that land (heck, the dictionary in my browser doesn’t even know it’s a word). The bottom line is that it’s going to be a long time before anything faster than Gigabit makes it to most desktops. So, I’m distinctly aware of the possibility of needing a manageable switch and eventually implementing 802.1ad link aggregation to increase bandwidth to the backup server and other key machines.

The major features I’m interested in:

  • 19″ rack mount form factor, at least 16 ports (24 would be good) all at Gigabit speeds.
  • Something from a good manufacturer, not a white-box or house brand. Good support.
  • Extensive manageability – at a minimum, SNMP and Telnet. Hopefully RMON.
  • 802.1Q VLANS
  • Good options for firmware upgrades, and not anything that requires a service contract (i.e. Cisco).
  • 802.3ad Link Aggregation (at a minimum, supporting 3 systems with 2-4 links each).
  • Port mirroring.
  • Support for QoS/CoS – at a minimum, 802.1p.
  • Acceptable MTBF.

Some added bonuses that I’d really like:

  • Support for syslog to a separate machine.
  • SFP slots so I can move to a faster backbone switch in the future, or move key systems to fiber or a faster standard.
  • Even more management – SSH would be nice, as would RADIUS or TATACS authentication.
  • A good switch fabric capacity – 32Gbps or higher for a 24-port.
  • 802.3af Power over Ethernet. (Well, I could always use injectors)
  • MAC filtering, ACLs, and/or 802.1x NAC.
  • 802.1ab LLDP.

I’ll keep looking around and pricing things out. My hard budget is $500, but it would take a lot to get me to spend that much. I’m probably looking for something closer to the $300 range, probably lightly used or refurbished. If anyone happens to see this and has a suggestion, feel free to share!