| Planning Stage | Project Plan Software Configuration Management Plan Software Quality Assurance Plan Software Project Risk Factor |
| Requirement Stage | User Requirement Specification Software Requirement Specification Requirement Traceability Matrix Form |
| Design Stage | Software Design Specification UML Designs |
| Development Stage | Source Code |
| Integration & Test Stage | Project Test Specification Test Plan Test case & conditions Test Strategies (Unit test plan, Integration test plan, System test plan, Validation test plan) Test Report |
| Installation & Acceptance | Deployment plan Release Note User Manual |
Friday, February 1, 2008
Internal Project Deliverables
Monday, November 12, 2007
Wonderful definitions of designations at office
1. Project Manager is a Person who thinks nine Women can deliver a baby in One month.
2. Developer is a Person who thinks it will take 18 months to deliver a Baby.3. Onsite Coordinator is one who thinks single Woman can deliver nine babies in one month.
4. Client is the one who doesn’t know why he wants a baby.
5. Marketing Manager is a person who thinks he can deliver a baby even if no man and woman are available.
6. Resource Optimization Team thinks they don’t Need a man or woman; They’ll produce a child with zero resources.
7. Documentation Team thinks they don’t care whether the child is delivered, they’ll just document 9 months.
8. Quality Auditor is the person who is never happy with the PROCESS to produce a baby. And lastly……………..
9. Tester is a person who always tells his wife that this is not the Right baby.
Meaning of HR
After 2 years of selfless service, a man realized that he has not been promoted, no transfer, no salary increase no commendation and that the Company is not doing any thing about it. So he decided to walk up to His HR Manager one morning and after exchanging greetings, he told his HR Manager his observation. The boss looked at him, laughed and asked him to sit down saying. My friend, you have not worked here for even one day.
The man was surprised to hear this, but the manager went on to explain.
Manager:- How many days are there in a year?
Man:- 365 days and some times 366
Manager:- how many hours make up a day?
Man:- 24 hours
Manager:- How long do you work in a day?
Man:- 8am to 4pm. i.e. 8 hours a day.
Manager:- So, what fraction of the day do you work in hours?
Man:- (He did some arithmetic and said 8/24 hours i.e. 1/3(one third)
Manager:- That is nice of you! What is one-third of 366 days?
Man:- 122 (1/3×366 = 122 in days)
Manager:- Do you come to work on weekends?
Man:- No sir
Manager:- How many days are there in a year that are weekends?
Man:- 52 Saturdays and 52 Sundays equals to 104 days
Manager:- Thanks for that. If you remove 104 days from 122 days, how many days do you now have?
Man:- 18 days.
Manager:- OK! I do give you 2 weeks sick leave every year. Now remove that14 days from the 18 days left. How many days do you have remaining?
Man:- 4 days
Manager:- Do you work on New Year day?
Man:- No sir!
Manager:- Do you come to work on workers day?
Man:- No sir!
Manager:- So how many days are left?
Man:- 2 days sir!
Manager:- Do you come to work on the (National holiday )?
Man:- No sir!
Manager:- So how many days are left?
Man:- 1 day sir!
Manager:- Do you work on Christmas day?
Man:- No sir!
Manager:- So how many days are left?
Man:- None sir!
Manager:- So, what are you claiming?
Man:- I have understood, Sir. I did not realise that I was stealing Company money all these days.
Moral - NEVER GO TO HR FOR HELP!!!
Have a Nice Day.
HR = HIGH RISK
Friday, October 19, 2007
Solid Business Plan
Getting into business is costly, and that is the bare truth. There are many entrepreneurs in this sea of people around us, but only with ideas in their head. Nothing in terms of actual businesses!
But then, they do not know where to look. If you see carefully, there are a lot of investors out there who are looking out for just some entrepreneurs like these – entrepreneurs, who have good ideas for business, but do not have the capital to make the required start. If you can manage to catch their eye, you will get the capital you want for your business, and then there is no saying to what heights you can reach.
But it is important to impress these investors. You can very well guess that there will be hundreds of thousands of entrepreneurs wanting to forward their business plans to investors, so that they can get the break in life that they want. Do all of them get it? Hardly! The success ratio for getting a business investment is about 1 in 1000, or even less, as the number of prospective business investment applications is piling up. So where do you stand a chance? The truth is, you have to have a very powerful business plan – something that will impress the investors so greatly that they will consider your application and keep it on top of the heap in their 'IN' tray.
If you strategize carefully, you can really achieve that singular glory. Getting considered by investors for starting your business is no mean achievement, and is possible only with long hours of work. And a power-packed business plan. But before you even start out with making a business plan, you must know about the different kinds of business investors out there. They are of two main kinds – the angel investors and the venture capitalists.
Angel Investors – Angel investors are usually individual people, or sometimes groups of people, who have so much wealth in their coffers that they have nothing better to do with it than take risks by investing in business. In the very least, they get someone to manage their money. At the most, they get a share in a future corporate business. Though the name suggests so, there is nothing divine about angel investors. They will want a share in your company, and will want a part of the profits too. But, since angel investors usually operate individually, there can be many differences here in the way they operate.
Venture Capitalists – Venture capitalists are the more common type of investors for businesses, since angel investors are few and far between. Venture capitalists, or VCs as they are endearingly called by potential business investment seekers, operate in forms of organizations. They adopt businesses of a special kind, and rarely venture out of the genre. They will also want high stakes of ownership in the business they are investing, and will want minute reports of progress. It is much more difficult to catch a VC's eye than an angel investor's. But your business plan can do the trick. Here are some expert ideas on making a very effective business plan that could catch the eye of even the most discerning of investors.
Tips on Making a Successful Business Plan
Tip # 1 – Be perfectly sure of what your business is set out to do
You will be surprised to know the large number of people out there who want to enter business and even have a germ of an idea in their mind, but do not know what exactly they want to do with their business when asked. Such are the people who fail terribly when they submit their proposals to the investors. VCs are highly trained to weed out such vague proposals. When you are making your business plan, begin with the Executive Summary. The whole success of your business plan will depend on this, because it is here that the initial impressions will be formed. In this section, you should mention clearly what the intent and purpose of your business is. Make a very clear mention of why the market needs your business, and who your target market will be. If the investor gets convinced there is a market out there for your business, it is half the battle won.
Tip # 2 – Make clear projections
It is actually funny to see sometimes how some entrepreneurs put in their profit projections in their business plans. Reading something like 'If we have about 1000 customers in the first three months, we could surely be touching be 50 million mark within the year' makes for interesting reading; but it is surely not going to hold any water with the ever-so skeptical investors. When you are making a projection, be realistic. How did you arrive at the conclusion that you will get 1000 customers in the first three months? What are your marketing plans for achieving that? Will you have production enough to meet the sales if they come up? Write clearly about all these factors. And avoid using conditional sentences that begin with 'if' in your projections.
Tip # 3 – Be realistic about the risks
There is no business without risks and venture capitalists know this only too well. When proposals come to them, they are the ones to first assess the risks of the proposals. So, you will do very well if you mention the risks beforehand in your business plan. That will give a realistic flair to your report. Otherwise, it will appear like a fantasy novel, nothing else. That will not go any good when your VC is considering all the potentialities of your business plan. In fact, you must keep a separate section, perhaps in the Marketing Analysis topic to mention all the risks that are associated with your business. Make it also a point to clearly elucidate what resources your business has, or will have, to counter these risks.
Tip # 4 – Keep your language simple
You must remember that your business plan is not a blueprint you are going to hand out to your engineers. This is a business plan, and it will be read by an investor. The investor, though quite wealthy, might not be very proficient at language. You must know that a large number of business plans are rejected just because the investors fail to understand what the proposer was getting at. So, you need to be very clear-cut in your language, and not use any difficult phrases. It is better to use points and bulleted formats wherever you can.
In conclusion, being frank and honest with your business plan always works. Come to the point directly, and do not beat around the bush. Investors have very little time, and chances are that they will only skim through your business plan. You have just about a few seconds to impress them into making an investment for your business. Keep that in mind, and you will probably be setting a date for inaugurating your business venture!
But then, they do not know where to look. If you see carefully, there are a lot of investors out there who are looking out for just some entrepreneurs like these – entrepreneurs, who have good ideas for business, but do not have the capital to make the required start. If you can manage to catch their eye, you will get the capital you want for your business, and then there is no saying to what heights you can reach.
But it is important to impress these investors. You can very well guess that there will be hundreds of thousands of entrepreneurs wanting to forward their business plans to investors, so that they can get the break in life that they want. Do all of them get it? Hardly! The success ratio for getting a business investment is about 1 in 1000, or even less, as the number of prospective business investment applications is piling up. So where do you stand a chance? The truth is, you have to have a very powerful business plan – something that will impress the investors so greatly that they will consider your application and keep it on top of the heap in their 'IN' tray.
If you strategize carefully, you can really achieve that singular glory. Getting considered by investors for starting your business is no mean achievement, and is possible only with long hours of work. And a power-packed business plan. But before you even start out with making a business plan, you must know about the different kinds of business investors out there. They are of two main kinds – the angel investors and the venture capitalists.
Angel Investors – Angel investors are usually individual people, or sometimes groups of people, who have so much wealth in their coffers that they have nothing better to do with it than take risks by investing in business. In the very least, they get someone to manage their money. At the most, they get a share in a future corporate business. Though the name suggests so, there is nothing divine about angel investors. They will want a share in your company, and will want a part of the profits too. But, since angel investors usually operate individually, there can be many differences here in the way they operate.
Venture Capitalists – Venture capitalists are the more common type of investors for businesses, since angel investors are few and far between. Venture capitalists, or VCs as they are endearingly called by potential business investment seekers, operate in forms of organizations. They adopt businesses of a special kind, and rarely venture out of the genre. They will also want high stakes of ownership in the business they are investing, and will want minute reports of progress. It is much more difficult to catch a VC's eye than an angel investor's. But your business plan can do the trick. Here are some expert ideas on making a very effective business plan that could catch the eye of even the most discerning of investors.
Tips on Making a Successful Business Plan
Tip # 1 – Be perfectly sure of what your business is set out to do
You will be surprised to know the large number of people out there who want to enter business and even have a germ of an idea in their mind, but do not know what exactly they want to do with their business when asked. Such are the people who fail terribly when they submit their proposals to the investors. VCs are highly trained to weed out such vague proposals. When you are making your business plan, begin with the Executive Summary. The whole success of your business plan will depend on this, because it is here that the initial impressions will be formed. In this section, you should mention clearly what the intent and purpose of your business is. Make a very clear mention of why the market needs your business, and who your target market will be. If the investor gets convinced there is a market out there for your business, it is half the battle won.
Tip # 2 – Make clear projections
It is actually funny to see sometimes how some entrepreneurs put in their profit projections in their business plans. Reading something like 'If we have about 1000 customers in the first three months, we could surely be touching be 50 million mark within the year' makes for interesting reading; but it is surely not going to hold any water with the ever-so skeptical investors. When you are making a projection, be realistic. How did you arrive at the conclusion that you will get 1000 customers in the first three months? What are your marketing plans for achieving that? Will you have production enough to meet the sales if they come up? Write clearly about all these factors. And avoid using conditional sentences that begin with 'if' in your projections.
Tip # 3 – Be realistic about the risks
There is no business without risks and venture capitalists know this only too well. When proposals come to them, they are the ones to first assess the risks of the proposals. So, you will do very well if you mention the risks beforehand in your business plan. That will give a realistic flair to your report. Otherwise, it will appear like a fantasy novel, nothing else. That will not go any good when your VC is considering all the potentialities of your business plan. In fact, you must keep a separate section, perhaps in the Marketing Analysis topic to mention all the risks that are associated with your business. Make it also a point to clearly elucidate what resources your business has, or will have, to counter these risks.
Tip # 4 – Keep your language simple
You must remember that your business plan is not a blueprint you are going to hand out to your engineers. This is a business plan, and it will be read by an investor. The investor, though quite wealthy, might not be very proficient at language. You must know that a large number of business plans are rejected just because the investors fail to understand what the proposer was getting at. So, you need to be very clear-cut in your language, and not use any difficult phrases. It is better to use points and bulleted formats wherever you can.
In conclusion, being frank and honest with your business plan always works. Come to the point directly, and do not beat around the bush. Investors have very little time, and chances are that they will only skim through your business plan. You have just about a few seconds to impress them into making an investment for your business. Keep that in mind, and you will probably be setting a date for inaugurating your business venture!
The 5 Most Common Mistakes Made By Startups
Entrepreneurs are no strangers to mistakes. Mistakes will happen - with considerable frequency - and the value in making those mistakes is learning from them and avoiding them in the future. You can also study the mistakes of others that came before. Plenty of successful entrepreneurs are quite open about mistakes they’ve made, why they made them and what they learned. We don’t need to keep repeating each other’s mistakes over and over.
But, that’s quite often the case. When it comes to startup mistakes you’ll see many companies making the same ones over and over.
Here are 5 of the most common mistakes made by startups:
- Staying in Stealth Mode Too Long. New startups seem quite fond of stealth mode (or its newer cousin “ninja mode”), when they’re hiding under the radar but still hyping just enough to try and pique interest. But stay in stealth mode too long and you run the risk of disappearing off the radar. Never mind the fact that you can’t sell your new product or service while in stealth mode and therefore can’t generate any revenue. There are plenty of reasons why startups launch too slowly; really you need to force yourself to launch and get past all the excuses.
- Not Focusing on the User. Who are you building your new product for? Who is the precise target? Many startups can give a generic answer to that question, but very few of them are really honed in on the specific wants of their “perfect user.” This is a combination of too little research and too much enthusiasm for what they think is “the next killer idea.” This mistake is compounded if you’re building something that you wouldn’t use yourself. Building something you would use makes things easier - you’re the target user. Otherwise you need to take a much more pragmatic approach.
As well, many startups take the approach of “being everything to everyone.” That strategy never works. You end up being nothing to anyone.
- Trying To Do Everything. If a task isn’t core to your business try and outsource it. Entrepreneurs are extremely fond of saying they wear many hats (which is true!) but there’s a limit to what’s reasonable in the hat-wearing department. Lots of things can be outsourced, and although you’ll be paying someone else to do the work, you’ll be freeing up precious time of your own. That time will be infinitely more valuable than the money you spend.
- Not Having Enough Infrastructure. Many startups don’t have the proper tools in place to start their business. Primarily, money and time. It’s getting cheaper and cheaper to start companies nowadays but it’s never free. Lots of people start companies without realizing how much money it’s actually going to take. When they clue in, and decide they don’t have the money to invest (or they’re not willing to part with it), they’re in trouble.
Startups face similar challenges with time. People often start companies while working full-time jobs. It’s doable but damn hard. And as soon as the startup gets a bit rocky or other interests come into play, the startup company gets shelved or delayed. Paul Graham comments on this beautifully in The 18 Mistakes That Kill Startups. His theory is that people get into startups half-heartedly and that’s what kills them. I think that’s part of the answer. The other side of that coin is that people truly do care and believe in what they’re doing, but they don’t have the infrastructure and bandwidth in place to make it happen.
Infrastructure issues are also related to a startup’s lack of connections and resources to find good vendors, good hires, mentors and people to rely on. A couple guys in a garage may have a great idea and tons of talent but when they need help securing a loan or handling a business-related task they may not have the network or foundation in place to support them.
- Forgetting About Branding, Marketing and Sales. I know there are examples of companies succeeding with a “build it and they will come” approach. Some people argue if you build something people want they’ll find it and plunk down their hard-earned money. It happens. But more often than not you need to develop real, actionable and savvy branding, marketing and sales strategies. You might have a great product and the wrong message. Or a killer software application that no one knows about. It’s rare to have a startup where the founders (or one of them) has real experience in branding, marketing and sales. The result is either all the founders do it (and often poorly) or they all pass the buck.
You can take a “build it and they will come” approach and hope for the world to pick up your scent and fall in love with you, or you can figure out how you’ll get the message out, what that message will be and how you’ll generate leads. Go with the latter.
The good news is that almost every mistake can be undone, and it’s rare that one mistake kills a startup completely. So feel free to make them - but skip those listed above…
Top 10 Steps for Success
Logically, every business owner or executive manager strives to succeed. For the benefit of the company, the employees, the product and last but not least, for themselves. Certainly, sometimes the order that i listed changes as little. 
I ran into a listing of the top 10 steps to succeed in business. Evaluate yourself, does your company do what's described below? Do you have room for improvement?
Here we go:
1. Develop a strategic plan. Many business owners measure their success by how hard they work and whether there is enough money to cover payroll, but in reality they operate best with a strategic plan and methodology for measuring and executing that plan.
2. Stay flexible to changes in the strategic plan. A strategic plan should be a work in progress that evolves in accordance with long term goals. Although flexibility is important, it does not mean that a business should be run without a strategic plan.
3. Transform yourself from company expert to master strategist. Focus on planning the future business success by creating a leadership team and delegating duties. This strategy might require an extensive role change, but will help to obtain more control for the future success of the company.
4. Focus on short-term growth. Key company goals should be monitored daily and weekly rather than monthly. By focusing on business growth in the short term there will be far less concern over the long term.
5. Develop reporting systems. Strategies cannot be implemented without reporting systems that track critical numbers. A daily review helps to measure and clarify where company efforts need to be enhanced as well as holds each employee accountable for performance.
6. Hold a daily management meeting. A daily meeting creates the intensity and focus needed for business owners to identify problems and issues before they get out of control. Meetings should focus on one key issue.
7. Control costs by budgeting percentages. Control costs by measuring a percentage rather than an absolute basis. Daily or weekly percentage increases and decreases should signal change and help to determine why inconsistencies might exist.
8. Offer incentives to key business drivers. All employees have the ability to drive or stall the business. Creative incentives that drive should contribute to both the profitability and mission of the company. Fair incentives should be tied to specific measurable items that each employee has control over.
9. Create a new management model. Creating a new management model allows for updating as well as preparing the business for positive change. Incorporate all levels of employees in both the thought and implementation process.
10. Play to win. These guidelines are easier and more enjoyable to adhere to when a company aims high and plays to win. A business that survives is the one that plans to innovate, allows for constant change and plays to win.
Found something that you're already doing? Anything that you're doing that leads to consequent success but is not mentioned here? Add it to the list, the more tips we can compile in this thread, the better it is for the community.
I ran into a listing of the top 10 steps to succeed in business. Evaluate yourself, does your company do what's described below? Do you have room for improvement? Here we go:
1. Develop a strategic plan. Many business owners measure their success by how hard they work and whether there is enough money to cover payroll, but in reality they operate best with a strategic plan and methodology for measuring and executing that plan.
2. Stay flexible to changes in the strategic plan. A strategic plan should be a work in progress that evolves in accordance with long term goals. Although flexibility is important, it does not mean that a business should be run without a strategic plan.
3. Transform yourself from company expert to master strategist. Focus on planning the future business success by creating a leadership team and delegating duties. This strategy might require an extensive role change, but will help to obtain more control for the future success of the company.
4. Focus on short-term growth. Key company goals should be monitored daily and weekly rather than monthly. By focusing on business growth in the short term there will be far less concern over the long term.
5. Develop reporting systems. Strategies cannot be implemented without reporting systems that track critical numbers. A daily review helps to measure and clarify where company efforts need to be enhanced as well as holds each employee accountable for performance.
6. Hold a daily management meeting. A daily meeting creates the intensity and focus needed for business owners to identify problems and issues before they get out of control. Meetings should focus on one key issue.
7. Control costs by budgeting percentages. Control costs by measuring a percentage rather than an absolute basis. Daily or weekly percentage increases and decreases should signal change and help to determine why inconsistencies might exist.
8. Offer incentives to key business drivers. All employees have the ability to drive or stall the business. Creative incentives that drive should contribute to both the profitability and mission of the company. Fair incentives should be tied to specific measurable items that each employee has control over.
9. Create a new management model. Creating a new management model allows for updating as well as preparing the business for positive change. Incorporate all levels of employees in both the thought and implementation process.
10. Play to win. These guidelines are easier and more enjoyable to adhere to when a company aims high and plays to win. A business that survives is the one that plans to innovate, allows for constant change and plays to win.
Found something that you're already doing? Anything that you're doing that leads to consequent success but is not mentioned here? Add it to the list, the more tips we can compile in this thread, the better it is for the community.
Thursday, February 15, 2007
ESRI's ArcWeb Services
ESRI's ArcWeb Services are a collection of tools that enable users to integrate mapping, routing, and geographic reporting into their Web sites, mobile devices, and desktop applications. As businesses grow and enterprise needs evolve, ArcWeb applications can be expanded without the need for additional tools or software reconfiguration. ArcWeb capabilities are built on ESRI's industry-leading GIS (Geographic Information System) software. ESRI, with decades of experience and proprietary knowledge in the GIS industry, provides sophisticated analytical GIS tools to solve spatial problems.
Developer resources are available to integrate ArcWeb services into any application. These services allow users to perform a variety of spatial query and proximity searches on a host of datasets provided by Tele Atlas, as well as on custom datasets uploaded and maintained by users. ArcWeb Services also allow users to create simple or multipoint routes, base maps and color-coded maps, and demographic reports. They also allow users to upload their own data, add it as a layer to an existing map service, and share it with other members of the enterprise.
Developer resources are available to integrate ArcWeb services into any application. These services allow users to perform a variety of spatial query and proximity searches on a host of datasets provided by Tele Atlas, as well as on custom datasets uploaded and maintained by users. ArcWeb Services also allow users to create simple or multipoint routes, base maps and color-coded maps, and demographic reports. They also allow users to upload their own data, add it as a layer to an existing map service, and share it with other members of the enterprise.
Tuesday, February 6, 2007
Serial UART, an introduction
An UART, universal asynchronous receiver / transmitter is responsible for performing the main task in serial communications with computers. The device changes incomming parallel information to serial data which can be sent on a communication line. A second UART can be used to receive the information. The UART performs all the tasks, timing, parity checking, etc. needed for the communication. The only extra devices attached are line driver chips capable of transforming the TTL level signals to line voltages and vice versa.
To use the UART in different environments, registers are accessible to set or review the communication parameters. Setable parameters are for example the communication speed, the type of parity check, and the way incomming information is signalled to the running software.
Serial UART types
Serial communication on PC compatibles started with the 8250 UART in the IBM XT. In the years after, new family members were introduced like the 8250A and 8250B revisions and the 16450. The last one was first implemented in the AT. The higher bus speed in this computer could not be reached by the 8250 series. The differences between these first UART series were rather minor. The most important property changed with each new release was the maximum allowed speed at the processor bus side.
The 16450 was capable of handling a communication speed of 38.4 kbs without problems. The demand for higher speeds led to the development of newer series which would be able to release the main processor from some of its tasks. The main problem with the original series was the need to perform a software action for each single byte to transmit or receive. To overcome this problem, the 16550 was released which contained two on-board FIFO buffers, each capable of storing 16 bytes. One buffer for incomming, and one buffer for outgoing bytes.
A marvellous idea, but it didn't work out that way. The 16550 chip contained a firmware bug which made it impossible to use the buffers. The 16550A which appeared soon after was the first UART which was able to use its FIFO buffers. This made it possible to increase maximum reliable communication speeds to 115.2 kbs. This speed was necessary to use effectively modems with on-board compression. A further enhancment introduced with the 16550 was the ablity to use DMA, direct memory access for the data transfer. Two pins were redefined for this purpose. DMA transfer is not used with most applications. Only special serial I/O boards with a high number of ports contain sometimes the necessary extra circuitry to make this feature work.
The 16550A is the most common UART at this moment. Newer versions are under development, including the 16650 which contains two 32 byte FIFO's and on board support for software flow control. Texas Instruments is developing the 16750 which contains 64 byte FIFO's.
What is NMEA?
The National Marine Electronics Association is dedicated to the education and advancement of the marine electronics industry and the market which it serves.
It is a non-profit association composed of manufacturers, distributors, dealers, educational institutions, and others interested in peripheral marine electronics occupations (quoted from a promo in "NMEA News")
What is an NMEA standard?
For the purposes of this article, an NMEA standard defines an electrical interface and data protocol for communications between marine instrumentation. (They may also have standards for other things.)
More about NMEA?
NMEA 0183 (or NMEA for short) is a combined electrical and data specification for communication between marine electronics and also, more generally, GPS receivers.
The NMEA 0183 protocol is a means by which marine instruments and also most GPS receivers can communicate with each other. It has been defined by, and is controlled by, the
The NMEA 0183 standard uses a simple ASCII, serial communications protocol that defines how data is transmitted in a "sentence" from one "talker" to one or more "listeners". The standard also defines the contents of each sentence (message) type so that all listeners can parse messages accurately:
• Each message starting character is a dollar sign.
• The next first five characters identify the type of message.
• All data fields that follow are comma-delimited.
• The first character that immediately follows the last data field character is an asterisk.
• The asterisk is immediately followed by a two-digit checksum.
NMEA is a standard protocol, use by GPS receivers to transmit data. NMEA output is EIA-422A but for most purposes you can consider it RS-232 compatible. Use 4800 bps, 8 data bits, no parity and one stop bit (8N1). NMEA 0183 sentences are all ASCII. Each sentence begins with a dollar sign ($) and ends with a carriage return linefeed (
Friday, February 2, 2007
Unified Modeling: What, Why & How?
Executive Summary
The use of Unified Modeling Language (UML) has been taken as one of the most important parts of software developments in the last couple of years. It is supposed to be a big step towards sorting out silly conflicts over notation. It’s important to know that creating a UML diagram involves a cost, and a UML model is not something that fundamentally matters to a customer. Customers are always interested in the working software that meets its requirements, not pictures & models. Hence the cost of creating a UML model for the system is always bearable by the developing organization only. Yet it is important to avoid the silly mistakes in system analysis and to save resources & time by avoiding conflicts & inconsistencies. Being UML modeling so important, one needs to know the focus of modeling & how to meet this focus. This collection of words includes about how to move with UML for system analysis & deployment before jumping into its development.
What's a Model:
A model may be taken as a form of abstraction that omits undesired details and allows the designer to look into the more important aspects and elaborate them with proper consideration. A model allows a designer to look into all the aspects of the systems so that no important phase or point or detail is left without discussion and consideration that may cause failure of development in later stages where the remedy actions are either impractical or they are too costly & time consuming. A model is used for understanding a system properly before its actual implementation so that the product may be developed in a cost effective and efficient manner. Also the systems developed with proper understanding are much more maintainable and feed to user requirements for long time.
To build complex systems, they need to be simplified first so that they may be properly understood & then developed in a maintainable way. For this the designer needs to watch the system from different focus points with abstract details. These focus-points give different “Views” of the system. These views of the system give a “Model” of the system.
Why Modeling?
While designing a system, we need to analyze and understand it from different angles & view points. So many different conceptual, logical, physical and other types of complexities need to be solved. For this we use different design strategies and models. Modeling creates a base for implementation of the system and its testing. It gives specifications for its use by customer. It explains the design idea in a presentable form. A Model explains the system at different levels of abstraction. At a time, it provides a selective aspect study of the system. The designer can use it to explain the architecture to the professionals and abstract and relevant details to others. Modeling it a strong tool to plan the steps in development of the system and check the feasibility of ideas before final actions on it. Modeling also provides a base for testing and testability of the system before the system is actually implemented. Moreover, modeling reduces the complexity of the system and hence saves resources.
It is important to mention here that the design is only a concept that describes the system understanding of the designer. It tells the thinking of an experience-holder about the strategy to be adopted for development of the system in hand. Now all human does not think in the same way. Some may have better understanding in textual form while others may prefer graphic notations of the design. Here the model efforts to bring them on an almost common platform where the conflicts of the design may be resolved and some consistent design for implementation may be proposed. Model is also important to remove the silly mistakes while developing the system and also creates a scale for evaluation while development. Its output is something more than a simple software system with its better understanding, documentation and maintenance tips.
OOMD & UML
Object-Oriented Modeling & Designing (OOMD) is an innovation for Inception, Elaboration, Construction and Transition in Software Development Life Cycle (SDLC). This is a way for analyzing, designing and developing a software system using concepts having direct correspondence with the real world entities. We use graphic notations for Object-Oriented construction of systems. Visual Modeling is one way for generating these notations. Visual modeling creates Object-Oriented-Models (OOM) for a system under construction for its analysis, designing & implementation in an object-oriented language. We create OOMs for complex systems under construction for their better understanding and a better & efficient designing. We use different graphical notations with lots of documentations for a proper & unambiguous understanding of the system. For such modeling of the systems, we can use a globally accepted media known as Unified Modeling Language (UML). UML gives a set of graphical notations prescribed for different entities & terms used in Object Oriented Modeling of any system.
How to start: Static Analysis
The system development life cycle starts with a problem statement using which a use-case model is created which is important to state the understandability & usability of the system. Here the system is represented at the highest level of abstraction i.e. (generally) as a single unit that can be used for some purpose. It includes all possible conversations (termed as use cases) between a user (termed as actor) and an operator in the system during any task. Sequence of occurrence of the dialogues is not important to be mentioned in a use case model. Also it is not required to show how the system is performing its task. It only shows how a system interacts with its environment/user.
The Next Step
Once the system & its interaction with its environment are understood, the analysis part of system development life cycle (SDLC) is almost complete. The next step is for designing the system. This design is later implemented in terms of some high level codes, which are then to be converted into executable modules and installed in the environment of the system. Hence designing creates a base framework to move forward towards implementation. In unified modeling, class diagrams are created for this step of SDLC. The appropriate classes can be found from problem statements from the list of nouns used in the problem statement. These classes interact with each other in terms of message passing. The classes create templates for objects in the system and hence we need to show the attributes as well as methods involved in the template clearly. Class diagram creates a base for implementation of the system in any Object-Oriented language. Lacking a proper description of types or interactions of classes in the class-diagram not only can weaken the base of quality of the system but can also introduce many inconsistencies in the system. So we need to clearly describe all required templates and interactions between them. For more clarity this analysis is done on different levels of abstractions from top to bottom.
Dynamic Analysis
Static analysis is followed by analysis of dynamic behavior of the system. Here the unified-model members include sequence diagrams & state diagrams. For software implementation point of view, the state diagram may not be of direct use but they serve for understanding the system dynamicity in terms of some finite number of states. This finite state analysis helps the developer to avoid many inconsistencies and exceptions at run-time. The state diagram clearly state the transitions of states on occurrence of some event or message passing between two classes. State diagram is highly important in concurrent processing or multithreading system designing & enhancing their performance.
Next in System Dynamics
System dynamic analysis includes another important analysis what we call ‘sequence analysis’. Sequence analysis talks about sequence of occurrence of events & operations. Sequence analysis gives direct information for implementation of a function or method. While creating a sequence diagram for a task or some function, it should be kept in mind that sequence analysis might be taken as the boundary of designing & implementation phases of SDLC from where we might be able to see both sides of the boundary. This means that a sequence diagram should be able to describe the direct mapping for the design to its coding in the programming language. Hence the sequence diagram can be understood like a graphical representation of the algorithm to implement a task in the system. The sequence analysis is important for each and every task. At the lowest level of abstraction, all the functions and methods included in a class are analyzed for occurrence of sequences of operations in it. This sequence can then be taken as it is and implemented in the chosen high level (Object-Oriented) programming language for the required function. The best sequence diagrams are expected to describe all the required programming instructions in the form of message-passing events.
Unified Modeling for System Deployment
After static & dynamic analysis and designing of the system, the software is implemented for the system in programming language. After that, the system needs to be installed in its working-environment including other interacting software, hardware and user interfaces. To analyze it before system implementation, deployment views in UML are used to show other interfaces, software & hardware, communication media and communication buses present in the environment. In deployment diagram, the information about the type of communication buses, server configuration, database architectures, type of interfaces with other used hardware and software etc. should be clearly mentioned.
Closure
No doubt that designing is one of the most crucial and resource consuming phase in SDLC and that’s too the one which has nothing of direct interest of the consumer. The developing organization itself is expected to bear the net cost of designing & modeling. In such a scenario, Unified Modeling has rapidly become the standard notation for modeling software-intensive systems. Companies today are interested in the entire project development life cycle and how it ties into the Unified Modeling. In most cases, the various methodologies will easily adapt to it. Unified Modeling provides a complete modeling at almost every stage of SDLC right from the beginning of requirement analysis till deployment of the system covering implementation model and dynamic behavior model also. For those of us who have built software and designed databases for many years, unified modeling will also remove additional risk from the development process. Although the unified modeling is not a SDLC process, it offers a concise and workable notation that can be applied to any domain.
Subscribe to:
Posts (Atom)