Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Comment Don't use software (Score 1) 164

A few people have said it already, but I'd like to reiterate:

Don't expect software to be able to do this for you. Tax laws change every year, both federal and state. Requirements for corporate filings change every year. Payroll requirements change every year (withholding amounts for FICA, FUTA, unemployment, health insurance requirements, etc).

I have an S-corp, but I don't do any of this myself.

I have a payroll service that costs $50 or so per month. For that cost, they do direct deposit, tax withholding, and also file quarterly 941 forms and prepare W-2s at the end of the year. All deductions are calculated by them, and their guarantee is that if they make a mistake, they take liability for it - ie, they'll pay any penalties or interest that arise as a result.

I have a lawyer who charges less than $200/year to prepare corporate filings. Additionally, they provide me with a list of changes to corporate laws over the course of the year. They highlight anything that may be important for me as a small business owner.

I have a CPA who prepares the corporate tax return. I do the day-to-day accounting (I happen to have Peachtree for Windows, I couldn't find anything open source or Linux based in 2000 when I started this business), but the CPA checks things over and prepares the corporate return and shareholder K-1 forms. This costs about $1000/year.

Overall, I pay roughly $2000/year for all of these services. I think this is well worth it, since I not only don't have to spend the time actually doing this work, but I also don't have to spend the time learning how I need to do it differently every year. I also get some measure of effective insurance, at least with the payroll company, since they are responsible for any mistakes they might make (they haven't made any yet).

I note that most of the question was really directed at legal questions. You can certainly get boilerplate contracts from several places. You may want to find a local lawyer and have them write you one. Get a confidentiality agreement, a subcontractor agreement, and a contract for services (that you provide). Once you have those, you can essentially reuse them for future contracts. There aren't a lot of requirements for meeting minutes (though one thing you should watch out for is that you are required to send a notice of shareholder meetings some 30 days before they're scheduled, or you need to have a "waiver of notice" form signed by all the shareholders. Of course this may not be true in your state) - just a list of people present, a very simple list of topics covered (summaries of the conversation are good, but are not necessary), and the dated signatures of the corporate secretary and president, who in my state can't be the same person.

The bottom line is that there is a lot of expertise needed to do these things right and keep up with all the annual changes. The cost of hiring experts is not that high, so it should be pretty close to a no-brainer.

- Steve

Comment Re:Good for them (Score 1) 111

There's a big difference between admitting to a bug that you can fix with a low/no-cost firmware upgrade, and admitting to a bug which requires a massive recall, and announcing to the market you'll be charging them more for '486 chips until you pay for the replacements.

There, fixed that for you.

Comment Re:Android is the Open Source replacement (Score 2, Interesting) 219

With the advent of Android on Linux, OpenMoko can safely retire. There will be a flood of Android hardware out soon in addition to the G1 and at least some of it will be hackable or open enough for developers to delve into the stack if they want. For example, you'll be able to improve the hardware drivers, add functionality left out by the original makers because they feared patent infringement, and take advantage of hardware acceleration that didn't make it into the shipping product. Perhaps the only sacrosanct portion kept off limits will be the radio stack itself, which if hacked could invalidate the CE mark, FCC, GCF, PTCRB, etc.

Android is software, not hardware. There is no guarantee that you will be able to write drivers, because not all manufacturers will give you datasheets without an NDA. There's no guarantee that you would get the source code to hardware drivers, since those can be non-GPL (resulting in a tainted kernel, but who cares, right?).

Unlike Android, OpenMoko is software and hardware.

You can also run Android on the OpenMoko hardware if you like (or Debian, or at least two other tailored distributions).

To the others asking, yes, the all the hardware is also open-source. You can download a pdf file with the schematics, and Pro/E models of the case. I asked Sean about "source" files for the schematics, and they haven't released them as yet. They're in Orcad, which is a multi-thousand dollar PCB design package. The gerber files for PCB manufacture are also not available (though I think they're in the pdf as well, so you'd at least get some good hints on layout there).

All software, including hardware emulators, is available online. Additionally, and what really sets them apart, is that you can get complete documentation from the manufacturers of every part they use - the processor, RF components, memory chips, etc. That's one thing that took a lot of time, and really restricted their design.

It's too bad Sean didn't mention the downsizing last week at the Embedded Systems Conference, but I guess that would have been pretty depressing to hear anyway.

- Steve

Comment Re:But if that's right... (Score 1) 774


The following is from a great book by A.K. Dewdney: Yes, We Have no Neutrons.

The formula is N = R* x Fp x Ne x Fl x Fi x Fc x L

For which:
R* = number of new stars that form in our galaxy each year
Fp = fraction of stars having planetary systems
Ne = average number of life-supporting planets per star
Fl = fraction of those planets on which life develops
Fi = fraction of life forms that become intelligent
Fc = fraction of intelligent beings that develop radio
L = average lifetime of a communicating society

The formula has appeared in several popular science magazines with the values set to:

N = 10 x 1 x 1 x 1 x 0.01 x 0.1 x L

So, N = 0.01 x L

The only numbers in the formula which anything other than a guess can be made are R* and L. Based on current observations most set R* at 10. Everything else in the formula would be a wild guess, except for L. More is known about L than any other part of the formula, since we are a communication society. Since we receive more and more of our communication from satellites, cable, and the internet, we are broadcasting less and less away from the earth. In the near future we will likely go dark as a significant source of radio/broadcast signals capable of being detected from space. If we say that our source of signals is about 100 years, drop the 100 back into the formula and you get 1. That must be us.

You have a slight units error in your analysis.

R* x Fp x Ne results in the number of new life supporting planets *per year*.

Multiplying the next three terms (Fl x Fi x Fc) gives us the fraction (or probablility) of those planets that will eventually have intelligent life.

So, to get the number of coexisting societies per galaxy, you can more or less take the number of new societies per year and multiply by the number of years during which those societies will be able to detect each other (L). The number of new detectable societies (with your wildly optimistic probabilities :) ) = 1.

That's not 1 technological society ever, that's an average of one in existence for any given year, in our galaxy.

There can be many many civilizations that rise and fall, and never know each other, over the course of billions of years.

Operating Systems

Submission + - Next Windows to Be Redesigned for Multicore CPUs

eldavojohn writes: "A Microsoft executive announced that the next Windows will be fundamentally redesigned to handle the numerous cores of present and future processors. The article notes that with Vista, the 20 year old GDI/GDI+ model was completely rewritten. Has Microsoft finally learned not to persist limitations and bad features of operating systems through generations? It will be interesting to see how Microsoft tackles the race conditions and deadlocks that come with pervasively multi threaded software and in the past complicated attempts to utilize multiple CPUs (like BeOS). Do you think it's it a smart move to further complicate an operating system to take advantage of these cores or should Microsoft concentrate on utilizing a single core for Windows while the applications take advantage of (possibly) more resources?"

Submission + - Multi-touch: the future of GUIs?

malbrech writes: "Multi-touch, the wonder piece of Apple's iPhone may have a much larger future. An article in describes how the genious behind Multi-touch, Jeff Han, developed the technology for large screens and gives a glimpse into the amazing potential born by the touch paradigm. There is also a fantastic video showing the use of multi-touch in various applications. Yes, reminiscences of AI become alive."

Slashdot Top Deals

Systems programmers are the high priests of a low cult. -- R.S. Barton