Monthly Archive:May, 2010

Internet Taxation of Software-as-a-Service

Recently, several states have made attempts at expanding their taxation of out-of-state businesses who provide services or products to customers within the state. (See generally, the Tax Foundation Special Report No. 176, March 2010).

In many of the analyses I’ve read, folks have jumped straight into the state law analysis. But, unless and until federal law changes, there are constitutional limits on states’ rights to tax out of state Businesses

Federal Law

The Supreme Court of the United States has issued a long line of cases which holds that in order for a state to tax a business conducted within that state there must be a “Substantial Nexus” between the business and the state.(*1) Developments in the delivery of electronic communications over the Internet have made it easier than ever before for out-of-state businesses to deliver goods or services to customers within states where they have no substantial nexus under the traditional test.

Specifically, the Supreme Court has issued a bright line distinction between . . . sellers with retail outlets, solicitors, or property within a State . . . on one hand and those who do no more than . . . communicate with customers in the State by mail or common carrier as part of a general interstate business . . . on the other hand.(*2) The Court has consistently held that businesses belonging to the second group (e.g. those who have no agents within the state, but communicate with customers and deliver products to customers via generally available distribution channels within a state as part of a general interstate business) may not be taxed by the state where customers reside because it places an undue burden on interstate commerce.

This initial federal legal analysis is very important to complete before performing the analysis of the applicability of a state’s tax law.

State taxation of goods and services that are provided by out-of-state businesses over the Internet is an evolving area of the law. In 2007, the U.S. Congress extended the Internet Tax Moratorium until the year 2014,(*3) signaling Congress’s commitment to prohibiting multiple and discriminatory taxes on Internet usage. Recently, several states have taken aggressive stances attempting to assert the right to tax goods and services delivered to such states via Internet usage., in particular, is actively disputing several of these newly enacted tax laws. Amazon has responded to laws that claim the state has a right to assert taxes on sales to residents in the state as a result of Amazon’s affiliate program by (i) canceling the affiliate program in the applicable state; or (*4) (ii) challenging the state’s right to tax it in court (and thereby subjecting the state’s tax collections to dispute and making them difficult for the state to use).(*5)

The case law that will arise as a result of Internet-based companies disputing these state laws should provide some additional clarity. Additionally, it is important to note that it is the U.S. Supreme Court’s interpretation of the Congress’s exercise of its powers under Commerce Clause of the Constitution that provides mostt of the limits on how far states may extend their power to tax out of state businesses. It is not only future case law that may modify the law in this area — in the event that the U.S. Congress were to pass new legislation with an express position on interstate commerce and state taxation of out of state business over the Internet, the law would necessarily change.

Therefore, Software-as-a-Service providers need to be diligent about staying abreast of new developments in the law in these areas to ensure that they are in compliance with the current laws of the United States as well as the various states where they have customers.

(1)Quill Corp v. North Dakota, 504 U.S. 298 (1992).
(2)National Bellas Hess, Inc. v. Dept. of Rev. State of IL, 386 U.S. 753 (1967)
(3)Tax Foundation Special Report No. 176, March 2010
(4)(e.g. Colorado, North Carolina, and Rhode Island) Id.
(5)The New York trial court found for the State of New York, the case is currently on appeal to New York’s intermediate court, the New York Supreme Court, Appellate Division. Id.


I received a form license agreement from the other side (a Fortune 100 company) with this most excellent Y2K provision the other day:

Supplier represents and warrants that neither performance nor functionality of the services, Products or systems is or will be affected by dates prior to, during and after the year 2000.

It’s not like I can strike it and argue that it shouldn’t apply.

But, really?

The Latest Case Against Facebook

On May 5, 2010, The Electronic Privacy Information Center (EPIC) filed a complaint with the FTC regarding Facebook’s privacy practices (or lack thereof).

The biggest two complaints, to my reading are that (1) Facebook unilaterally tried to convert some information previously designated as private to public; and (2) Facebook changed its developer data retention policy to allow developers to retain end user data indefinintely.

Neither of these changes benefits end users, no doubt. But, what I’m fascinated to see is that today, a mere 12 days after the complaint, the user experience is significantly different from the experience described in the complaint (notably, the experience is more protective of user’s data when compared against the experience described in the complaint).

The legal process is slow and cumbersome and using it to argue with a quick and nimble internet-based adversary is going to be frustrating, to say the least. However, where end users are concerned, perhaps the quick responsiveness of Facebook is a benefit. If enough people complain, they just roll out a fix, long before the Feds, or the courts order them to do so. Certainly, this means that the fix is likely to be on Facebook’s preferred terms, rather than what the court or Feds order, but isn’t a quick fix better than a long period of open sharing without a fix (when it comes to privacy)?

I’m not saying I approve of Facebook’s most recent blunders. But, I do applaud of their quick “opt-in” and “opt-out-of-all” additions after the complaint about the blunders. And, I’m fascinated to see how or where the law fits in this world where the facts upon which any legal claims may be based are so ephemeral.

The Google Shareholder Meeting: Tidbits

Yesterday, I took some time out of my day to attend the Google annual shareholder meeting.

This meeting is more of a formality than many such meetings held by public companies.  As of the record date, the officers and directors held a total of 70.2% of the voting power of the company — so, obviously, everything has (or should have) been decided before the meeting.

Perhaps because of this, the formal portion of the meeting was a dry speed-reading session of various folks reading their portion of the Proxy Statement out loud.

In contrast, the complimentary on-campus lunch on the patio and the product demos prior were a big hit with those in attendance, as was the informal presentation and post-vote question and answer period.

Eric Schmidt’s presentation included a great Chrome ad, and some fun Internet facts, such as:

-10 years ago, there were 300M users on the Internet, today there are 1.2B.

-Today, there are 800 exabytes (1 EB = a billion GB) of information on the reachable Internet.

-Every minute there are 24 hours of video uploaded to YouTube and about 1/2 of those videos receive comments

-Eric said, “If you’re not using Chrome, you need to try it — everyone else is starting to use it.”  I chuckled internally when I heard this, since I was sent to building 43 to print my shareholder proof, and the browser they presented to me was Firefox.

-Google Translate is translating 160M pages/day.  And Larry, when asked by the man behind, said that he thought translate was the Next Big Thing. (Note — for more info check out this Play-by-Play post on the meeting)

Other than that, several folks took the microphone to give heartfelt praise and thanks to Google for their stance in China, and several others took the microphone to denounce and complain about the horrid handling of the proxy materials (Eric Schmidt asked  Patrick Pichette to personally meet with each of the grumpy folks after the meeting — I bet that was fun).

My favorite microphone participant was the very excited woman from Frederick, MD, who drove all the way across the country in her Google-themed car (with a brief stop in Topeka) to make a personal plea for Frederick, MD (with a second request on behalf of Topeka) to win the Google Fiber-Optic City contest.

All-in-all, it was a great way to spend a couple of hours.  Given that it’s down the street from my office, I think I may take the afternoon off to attend next year as well.

It Begins

As a gardener and tomato lover, my favorite garden season is when we get to watch and enjoy the fruits of the Summer garden.

Over the last few weekends, we’ve removed the remnants of the Spring/Winter garden, amended the beds, turned the soil, added an entirely new raised bed (thanks to the husband for the manual labor on that one), rejiggered the watering system, and transplanted and seeded for the Summer’s bounty.

The final result?


That’s 29 varieties of tomatoes, 4 hot peppers, 5 eggplants, 10 or so okra plants, 1 butternut squash, 3 cucumbers, 1 yellow crook-neck squash, leeks, shallots, garlic, and nasturtiums, and borage to bring the bees. We’re also trying our hand at some lettuces grown in the shade — typically, it’s too hot and they go to seed, but we can always keep experimenting.

The majority of the seeds we planted sprouted (except the yellow squash, so I bought some seedlings at the nursery and will be putting those in today). One new addition that we’re growing from seed is red okra. Historically, we’ve grown green okra, and loved it. But this year it will be growing side by side along its red cousin:


Right now, they are almost impossible to tell apart, but at some point, they will be quite different.

And finally, we have our aging herb box (that really needs a thorough re-do) and 5 varieties of basil (which will be hilariously overgrown in a month or so):


The Real Risks of Open Source Software

Every software start-up company I’ve ever worked with uses (or did use) some form of open source software. And yet, high level executives and board members at many of these companies, when asked whether their company uses any open source software, would regularly answer, “No” without hesitation.

Where is this disconnect coming from? Open Source Software is often perceived as “risky” or “untested” or “a liability nightmare” or, in the worst case, “an infectious disease” by some business folks, while most technical software people believe the correct use of open source software to carry minimal risk.


There are risks associated with using any third party’s software. When that third party is unidentified, not bound by a support agreement, based out of a foreign country, and/or impossible to get a hold of, then yes, it is fair to say that using it would be “risky” when compared with an established company with a business reputation to protect and an SLA to cover errors.


In some case — it’s true. There are many untested open source projects out there. These tend to be associated with a handful of developers instead of an active community, and a little due diligence should be able to help a start-up understand whether this particular ill is a problem with the open source software they are considering using.

A Liability Nightmare?

This is, by far, the most complicated issue I face as an attorney who deals with open source issues. At its most basic, the liability equation associated with open source software is the same as that associated with any third party component. The third party would like to disclaim liability for your use of their product.

The benefit of many proprietary software licenses is that the licensor may provide limited coverage for Intellectual Property claims related to their software. But, most software publishers go to great lengths to limit the amount and type of liability they will cover relating to a third party’s use of their software.

The typical open source license expressly disclaims all liability associated with the use of the software — effectively, it comes “AS IS” on a pure “BUYER/USER BEWARE” basis. On the other hand, if you read the license agreements of proprietary software carefully, you will find that most software (unless you pay quite a bit for it), comes with an express limitation on liability that is on the order of magnitude of the purchase price. It is consistent with the approach taken by proprietary software publishers that open source authors are liable for damages related on the order of magnitude of the license fee they receive (e.g. $0).

An Infectious Disease?

One flavor of open source licenses places conditions of “freedom” upon the use of the licensed code. The most famous of these licenses are the GPL, LGPL, and the AGPL. Essentially, these licenses require, as a condition of some uses or distributions, that software code combined with code licensed under these licenses must also be made available under the same license.

These conditions make these types of licenses “viral” because they may extend the license terms to some of the additional code (e.g. the start-up’s code) that the licensed code touches.

The key word in the previous sentence is *MAY.*

Actual Risk

The thoughtful evaluation of the issues outlined above and a comparison of the likely downside against the monetary benefit of using an open source component brings a start-up to understand what I call the “Actual Risk.”

By far, the most complicated part of the Actual Risk evaluation is the technical and legal analysis related to viral licenses. However, a technical read of the license by a knowledgeable tech attorney and code review with an engineer is likely to provide a good engineer or architect with comfort that the start-up’s use of a particular open source component is not subjecting the start-up’s code base (or the portion of the code base that they care about keeping proprietary) to any “viral” risk.

The Resource Risk

Investors and potential acquirors will want their own attorneys or possibly even code auditors to assess the Actual Risk, regardless of how correct the start-up’s own analysis may be. This investigation and analysis is a time and resource drain that can be minimized by good record keeping, but can never be entirely eliminated. Even in the event of zero Actual Risk, a company will incur some Resource Risk in connection with their use of open source software.

The FUD Risk

No matter what the final conclusion may be after the Resource Risk and the Actual Risk have been assessed and assumed by a start-up, there is the risk associated with the fact that a board member or a CEO will have to answer “Yes” to the question “Do your products contain open source?” A board member or CEO may not have the time to understand the outcomes and analysis of the folks who have willingly taken on the Resource Risk and the Actual Risk. If challenged, a board member or CEO need to feel confident that they can answer the question honestly, without incurring undue scrutiny or concern. In my opinion, the biggest risk associated with the use of open source software (assuming there is no Actual Risk that hurts the start-up’s business) is the FUD risk.

The best way to combat the FUD risk is to educate board members and CEOs so that they can comfortably speak about the company’s intelligent use of open source software as a cost reduction tool in areas where the Actual Risk is minimal or non-existent and the Resource Risks are less than the costs of the proprietary alternatives.