Edit: The answer may be a bit unclear if you don't have much background in economics, so here the tl;dr version.
- In the market for a specific kind of software (like an operating system) there will always emerge a monopolist, unless you have some heavy regulation from the start (this is explained by positive externalities from the network effect, and a lock-in causing the network effect). It isn't clear who of the initial players will emerge, but he will be able to do so without any unfair business practices, because the customers will flock to him - first for some unknown reason (or one not explainable with the classic rational actor theory), then, after he gains some critical mass, just because he is bigger than the competition. Microsoft happened to emerge in the OS market. They didn't need to abuse a monopoly position for that. The correct term is "natural monopoly".
- The lock-in effects result in a welfare allocation which is heavily oriented towards the supplying actor and puts the demanding actor at disadvantage. It could be argued that a company which builds stronger lock-in effects is doing "abuse" in an everyday sense (in classic microeconomics, the actual allocation doesn't matter, as long as it is efficient. There are other branches of economy which try to achieve fairness. There are 3-4 mutually exclusive definitions of fairness used in them). But both the building-dependency practices and monopoly are results from a market which allows a lock-in effect, the monopoly isn't caused by the practices.
Ergo, the success of Microsoft is not caused by its monopoly position. It is caused by unconventional qualities of the goods they produce, and the resulting market structure. They did some things which weren't nice for their customers or competitors, but these weren't result from being a monopoly. For more explanation, please read the original answer below.
"Monopoly" is a term stemming from economics, especially microeconomics. I'd like to give you some economical analysis here, because your question is somewhat inconsistent from an economical point of view.
The effect of a supply monopoly is that it is selling its products at a price which equals marginal revenue (instead of the usual marginal costs). This means that the supplier is selling less units at a higher price. It makes a bigger profit than under competition, but there are less people who buy the product, so the sum of the welfare over all marketplace actors is less than in the competition situation.
This is by no means considered abuse, just rational behavior for a monopolist. It is only a market failure if the good produced is deemed to be important for social instead of economical reasons, so it is desirable to be more widely available. For example, transportation is considered very important, so it is subsidized, and the presence of airline alliances is caused by the refusal of governments to allow more merger of airlines.
While there is some effort made to prevent monopolies from forming, there are sometimes benefits from their existence. A monopoly's products compete only against earlier products of the same company. This is a powerful innovation driver (because they can't compete on brand image, service level, etc. against their own products), especially effective when combined with the fact that with their big market share gives them lots of capital to invest in R&D.
This means that monopolies aren't good or evil. They are just another form of market structure, which has shortcomings (reduced welfare) as well as advantages. Selling at a higher price is not "abuse" in the sense used by politicians.
But note that we are not talking about any market here. We are talking about selling software, and this is a market with some quite unusual qualities.
First, the point of selling at a higher price when you are a monopoly is moot in this case. For information products, nobody ever sold at marginal costs - they were too low to reflect the value of the product in the analog age, and they are practically zero in the digital age (the marginal cost is defined to the cost of creating another unit of your product after you already have created the first n units). (OK, there are companies who do it - that's why you get so much digital products for free. But they are never pure producers, they are intermediaries which act in a very complicated market - Google gives you quality search for free, but bundles it with selling your information to advertisers for money. They are producer on the search market and intermediary on the second). So the monopoly is charging what the market can bear, and the non-monopoly is doing the same thing. How is that to be considered abuse?
There is another peculiarity which a software market shares with a few other markets and results in typical phenomena, including monopolies. It understandable that you confuse its effects with the effects of having a monopoly in the market. Older economical theories just noted that there are such strange cases where a market doesn't have the usual negative economies of scale (caused by excessive overhead), but that it can have positive economies of scale. As this happened in a very few markets, and ones which were heavily regulated and subsidized anyway, it wasn't such a hot research topic as it became in the days of ubiquitous information technology.
Today, it is established that the positive economies of scale are mainly caused by positive externalities, mainly the network effect. If you are on this site, you probably have enough background in information theory to have heard that the value of a network is proportional not to its nodes number, but to its nodes number squared (IT specialists call this Metcalfe's law, the Wikipedia article on it is good). This leads to the effect that the bigger a network is, the bigger it tends to get (winner takes it all phenomenon). This is what causes monopolies in markets like transport and telecommunications. While there is no network as such in software (I am talking application software here, not networking software at the low TCP/IP levels), there is another peculiar quality of software markets which creates a virtual network: The lock-in effect.
The lock-in effect says that once you have decided that you use one supplier, you can't switch that supplier without a really high cost. This is true for software - back in the old days, you practically couldn't exchange files between Microsoft and Unix systems without major conversion issues - but also for a number of other products (print cartridges, DSLR lenses). This creates a quasi-network: your Word 97 files are much more valuable if there are more people in the world who use Word 97 instead of competing non-compatible word processors. Such lock-in effects are what you are referring to in your question body. They are not a result from the monopoly; they are its cause. So while they are quite bad for the customer, they are by no means an abuse of monopoly.
There is a very easy way to remove such effects: Just ensure compatibility between competing products. Obviously, a company could do it, but very few of them have the interest to do so. A startup in a network market does have the interest, because else they can't gain a market share at all. This is why OpenOffice could always read and write MSOffice formats, and MS Office was slow in adding support for OO formats. There are very few exceptions where a firm in a networked market is able to hold a share with a non-compatible product without being crowded out; Apple is such an example, but it is based on a completely unrelated economic effect called Veblen good, plus some differentiation (Wintel wasn't a direct competitor in the publishing sector).
So monopoly (neither its abuse, nor its mere existence) definitely didn't cause Microsoft's success. The uncooperative business practices you mention were part of the success - it would have had less sales if it had removed the lock-in effects through embracing open standards - but I wouldn't call that abuse, just rational economical conduct. If society does not wish such a result, then it has to be observed as market failure, not as abuse, and the government should fulfil its duty of economic regulation on a failure-plagued market and impose regulation in order to change the result. (That's what the antimonopoly process against Microsoft was). Or they may let the market run its course, if the effects are not too bad for the overall welfare or for non-economic goal variables. Remember, the positive externalities create a somewhat unfair situation where the winning producer has a disproportionately big piece of the pie, the customers have a disproportionately small one, and the competitors get almost nothing - but this is not an especially bad effect in economics, mere allocation inefficiencies are considered minor there. But they also create a much bigger pie, so the general public may settle for this rather than for a heavily regulated market with a more (socially, not economically) fair allocation but less welfare overall.
If you want to know more about the economical analysis of information markets, read Information Rules by Shapiro and Varian. If you need a text on basic microeconomics (it applies to most of the "new economy"), the same Varian has written a very popular textbook on that. If you are interested in analysis of more complex markets, including but not restricted to the markets for information goods, try Market microstructure by Spulber and An introduction to the economics of information by Perez and Macho-Stadler. These are my sources for the explanation above. NB: the last two books require at least intermediate level knowledge of microeconomics.