Fourth Strategy: Expansion Software and 8020 rumors

zhaozj2021-02-16  82

Author: Joel Spolsky Translation: Bo Yang edit: Wenjing Jiang March 23, 2001

In 1993, Microsoft's spreadsheet software Excel 5.0 was published. It is a huge software that requires 15 trillion hard disk space. I still remember that the PC hard disk of the earliest listed (around 1985) is only 20 megabytes, so there is a lot of 15 trillion.

When Excel 2000 is published, it takes 146 trillion hard disk space, almost ten times in 1993! Microsoft's programmer is really bad, right?

wrong.

I dare to bet, you think I want to write an annoying, criticizing the expansion software. This article is everywhere in the Internet. A batch, grievance, these huge bloated software, so that I hate, or VI and Edlin are much better than Microsoft Word and Emacs, because the first two is very small ...

Haha, let you be! I won't write such an article because that kind of saying is not right.

In 1993, according to the price of the hard drive at the time, Microsoft Excel 5.0 took up a hard disk space of $ 36.

In 2000, according to the price of the hard disk at the time, Microsoft Excel 2000 took up a hard disk space of $ 1.03.

(These numbers are calculated according to the hard disk history price data here, to calculate the effect of inflation.)

From the actual economic point of view, Excel seems to be smaller!

Say truth, what is expansion software? Jargon File defines it as: "Provides the lowest function, while occupying the software that is not proportional to the memory space. In particular, describe the application software and the operating system upgrade. This word is on Windows / NT. Common, Windows / NT is also its source. "

I guess this helper is just hated Windows. Since the appearance of Windows 386 (1989) since the virtual memory, I haven't finished more than ten years. Besides, the hard disk price has dropped to $ 0.0071, and the more falls.

Maybe Linus? Kerlund can explain this problem clear. He wrote on his webpage: "A big disadvantage of these expansion software is that even if you just want to do a small thing, you have to load this big program. It puts your memory full use. ... You don't use your system effectively. It is unnecessary to make your system efficiency very low. "

I understand, it turns out that it is used by your memory. Well, wrong. Since Windows 1.0 published in 1988, the operating system only places the used code page in memory. If you have a 15 trillor executable, but only two megabytes of code pages are running, the operating system only puts these two megabytes from the disk into the memory. And if you use a comparative version of Windows, the operating system will automatically adjust the hard drive, put the two megabytes page together, so that the program starts faster.

I think no one will be negative. On the powerful and cheap computer today, start a big program than starting a small program only five years ago. So what are those people call?

Ra Downes gives us a prompt. It seems that he spent a few hours and studied a small program of Microsoft, and because this program had a megaby size, he was very angry (he wrote that article, only 3.15 points worth 3.15 points). With him, that program should be 95% small. It is ridiculous that the program he studied is called regclean, you are likely to have never heard of it. Its role is to delete something that is useless in Windows Registry. If you are worried about something that is useless in Windows Registry all day, your character is likely to be fascinated by things, it is difficult to restrain the weaknesses. So I suspect that critical expansion software is a mental illness, not software issues. In fact, the existence of expansion software is reasonable. At the very least, if the programmer doesn't have to worry about the size of the software, the software can publish it earlier. Users can get more features earlier. These functions are beneficial when they are not used. If your software store spends two months before publishing software, it is less than 50% of its software, which is almost unable to do it. Maybe, if your hard drive is full, you can download a MP3 song more. But you spend more than two months, such as new software publishing, you will feel. And your software chamber will lose two months of sales, and the loss is greater.

Many people who engage in software development are trained by the old law of "80/20". This law seems to be very reasonable: 80% of people use only 20% of the function. So you think that you can get 80% of sales of 80% as long as you implement 20% of the function.

Unfortunately, that is not the same 20%. Everyone uses different features. In the past ten years, I have heard of dozens of companies, making it determination not to learn lessons, trying to publish "light shaped" versions, only 20% of the functional text processing software. Most of the situation is that they send the software to the reporter where to comment, those reporters' comments, use this new text processing software to write their comment articles. The article is over, the reporter will use the "word number", because most reporters have clear words limit when writing articles. But the function of the "word" can't be found in the software, because it is "80% of no one". As a result, the reporter will write an article. On the one hand, how is this "light shape" software, how is the expansion software? On the other hand, I can't use this software because it doesn't have the "word number". And this article often writes.

When you go to sell your "light shaped" software, you said to others: "Hey, this software is very small, only one megabytes." People listened usually very happy, then ask you if you have anything for them Very important feature, if not, you won't buy your software.

Basic Probably: If your strategy is "80/20", it is hard to sell your software. that is the truth. This strategy has never been victorious since the start of the software industry. Surprisingly, a lot of unparalleled company's senior executives still think it is a good idea.

Jamie Zawinski is best to speak when discussing the first version of Netscape that changes the world: "Mozilla [Netscape 1.0] is not because it is all used inside (if it is true, it is very convenient to explain). Mozilla is very large, Because you need a big need. You need a lot, because the Internet is a big thing. There are many small browsers on the Internet, which is basically useless. We never want to write Mozilla. Make it a perfect pearl. "

转载请注明原文地址:https://www.9cbs.com/read-13080.html

New Post(0)